Download:
pdf |
pdfMillennium Cohort Study:
Telephone Study of 3,000 Non-Responders
Final Report
Submitted to
Naval Health Research Center
DoD Center for Development Health Research
PO Box 85122
San Diego, CA 92186-5122
Submitted by
RTI International
3040 Cornwallis Road
Research Triangle Park, NC 27709-2194
RTI Project Number: 0209522.002
December 23, 2005
Table of Contents
1.
2.
3.
Overview of the Study............................................................................................................. 1
1.1
The Study Sample ......................................................................................................... 1
1.2
The Non-Response Study Questionnaire ...................................................................... 2
Data Collection Activities ....................................................................................................... 4
2.1
Interviewer Training ..................................................................................................... 4
2.2
Telephone Survey Quality Control Procedures ............................................................ 4
2.3
Tracing Results ............................................................................................................. 6
2.3.1
Batch Tracing.................................................................................................... 6
2.3.2
Intensive Tracing .............................................................................................. 7
2.4
Call Outcomes............................................................................................................... 9
2.5
Data Collection Outcomes .......................................................................................... 11
2.6
Data Collection Debriefing Summary ........................................................................ 13
2.6.1
Highlights from the Tracing Debriefing ......................................................... 13
2.6.2
Highlights from the Interviewer Debriefing ................................................... 14
Questionnaire Results and Recommendations ...................................................................... 16
3.1
Results and Initial Recommendations......................................................................... 16
3.2
Summary of Recommendations .................................................................................. 28
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
i
Final Report
December 23, 2005
List of Exhibits
Exhibit 2-1. Data Collection Schedule ..........................................................................................4
Exhibit 2-2. Batch Tracing Results ...............................................................................................7
Exhibit 2-3. Summary of Individual Submissions to Intensive Tracing .......................................8
Exhibit 2-4. Intensive Tracing Results ..........................................................................................9
Exhibit 2-5. Increase in the Number of Call Attempts by Date ..................................................10
Exhibit 2-6. Mean Call Count by Summary Status Group ..........................................................10
Exhibit 2-7. Percent of Completed Cases by Call Attempt .........................................................11
Exhibit 2-8. Final Dispositions....................................................................................................12
Exhibit 3-1. Main Reason for not Participating in this Round of the Study ...............................16
Exhibit 3-2. Second Reason for not Participating in this Round of the Study ............................17
Exhibit 3-3. Awareness of MilCohort Incentive Offer................................................................18
Exhibit 3-4. First Item Chosen to Encourage Participation.........................................................19
Exhibit 3-5. Second Item Chosen to Encourage Participation ....................................................20
Exhibit 3-6. Magazines Cited as Way to Encourage Participation .............................................21
Exhibit 3-7. Amount of Cash Suggested to Encourage Participation .........................................22
Exhibit 3-8. Health Status Reported by MilCohort Non-Response Study Participants ..............23
Exhibit 3-9. Percent of Non-Response Study Participants Who Indicated They Received
Mail or E-mail Survey Notifications .................................................................24
Exhibit 3-10. Internet Access of MilCohort Non-Response Study Participants ...........................25
Exhibit 3-11. Survey Notification Requests..................................................................................26
Exhibit 3-12. Updated Contact Information Provided by Participants Requesting Hard
Copy Questionnaire...........................................................................................27
List of Appendices
Appendix A – Questionnaire ...................................................................................................... A-1
Appendix B – Affidavit of Non-Disclosure................................................................................ B-1
Appendix C – Confidentiality Agreement .................................................................................. C-1
Appendix D – Debriefing Form.................................................................................................. D-1
Appendix E – Frequency List ......................................................................................................E-1
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
ii
Final Report
December 23, 2005
1. Overview of the Study
The Millennium Cohort Study (MilCohort) was designed to evaluate the long-term health
of enrolled members of active duty and guard/reserve military by following a cohort first
identified in 2001 for up to 21 years. MilCohort requires about 30 minutes of participant’s time
every three years for the follow-up interviews which can be completed either on the web or via a
mail survey. With an initial sample size of 77,000, MilCohort has the potential to be the
definitive source for evaluation of the long-term health of people who served in the military, and
to benefit future members of the military.
The success of MilCohort depends on retention of cohort members throughout the
follow-up rounds. The MilCohort project team is aiming for a 90 percent retention rate for this
cohort. However, after the first follow-up survey conducted from June 2004 to February 2005,
there were still approximately 29,000 non-respondents (or 38 percent). This prompted the Naval
Health Research Center (NHRC) to fund a study of 3,000 non-respondents to gather information
on why sample members are not participating in the current round of MilCohort and to prompt
sample members to participate.
This report describes the activities associated with conducting the Millennium Cohort
Study: Telephone Study of 3,000 Non-Responders (MilCohort Non-Response Study), which RTI
conducted for the NHRC. The pilot study consisted of telephone prompting calls to this subset of
cohort non-respondents, as well as moderate tracing of non-respondents that could not be
reached by telephone. The survey was administered to active and non-active military personnel
who had not responded to the current round of the full self-administered Millennium Cohort
Study. In the following sections, we describe the study sample and survey instrument. Chapter 2
provides an overview of the data collection methods, and Chapter 3 provides the results of the
non-response questions as well as suggestions for areas of improvement for future rounds of
MilCohort.
1.1
The Study Sample
NHRC provided RTI with a sample of 3,000 randomly selected cases of non-respondents
to the Panel 1 Millennium Cohort Study. The data file of non-responders contained the following
variables:
•
Social Security Number (SSN)
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
1
Final Report
December 23, 2005
•
Full Name
•
Date of Birth
•
Sex
•
ID Correspondence
•
Status Date
•
Bad Address Indicator
•
Complete Mailing Address
•
E-mail Address
•
Phone Number
After obtaining the sample file and prior to the start of data collection, RTI submitted the
sample file to outside vendors to perform batch tracing steps. These steps are discussed further in
Section 2.3. On a regular basis throughout data collection, RTI submitted updated contact
information (e-mail, address, telephone number) as provided by respondents to NHRC.
1.2
The Non-Response Study Questionnaire
The MilCohort Non-Response Study questionnaire was developed by a joint effort
between the MilCohort Study team and RTI. The non-response questionnaire covered five areas:
introduction, informed consent, non-response questions, participation prompting, and closing. A
copy of the final questionnaire can be found in Appendix A. The questionnaire was programmed
using Blaise software, and underwent several rounds of testing before data collection began.
A short introduction let sample members know that RTI staff were calling on behalf of
the U.S. Department of Defense (DoD). The informed consent provided the respondent with key
information which they could use to make an informed decision about participation in the nonresponse study. This information included the length of the survey, the protocol to share the
information obtained with the MilCohort Study team at DoD, confidentiality assurance, notice
that participation was voluntary and the participant could stop the interview at any time.
The non-response section was the main portion of the instrument. Questions asked in this
section included:
•
Reason for not participating in the current round of the MilCohort Study,
•
Awareness of incentives,
•
Evaluation of items to better encourage participation, and
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
2
Final Report
December 23, 2005
•
Reporting of health status.
Participation prompting followed the non-response section and allowed for collection of
e-mail addresses as well as updated contact information (including mailing address and
telephone number). The survey ended with a closing section thanking the respondent for their
participation in the non-response study.
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
3
Final Report
December 23, 2005
2. Data Collection Activities
NHRC provided a sample of 3,000 non-respondent cases that were sent to RTI’s Call
Center Services (CCS) unit. Interviewing began on September 8, 2005, and ended on November
11, 2005. Activities conducted prior to and during the data collection period are summarized in
Exhibit 2-1.
Exhibit 2-1. Data Collection Schedule
Activity
Batch Tracing
Telephone Interviewer Training
Training for Tracing Staff
Intensive Tracing
Data Collection
Dates
September 1-8, 2005
September 6-7, 2005
September 15, 2005
September 19, 2005 - October 27, 2005
September 8, 2005 - November 11, 2005
Detailed descriptions of the training, tracing, and data collection activities are provided in the
sections that follow.
2.1
Interviewer Training
We trained a total of 13 experienced telephone interviewers, 2 Team Leaders, 1
Supervisor, and 1 Monitor for the study. All telephone interviewers were required to attend a 6hour project-specific training session before starting work on the survey. The trainers used a
combination of lecture, role-play exercises, question and answer sessions, and demonstration and
practice interviews. The demonstration and practice interviews were designed to focus on
procedures involving particular situations that the interviewers might encounter when
administering the interview. A portion of the training was devoted to a discussion of the
background of the project, the longitudinal nature of the study design, and gaining cooperation.
Additionally, all interviewers, supervisors, and monitors were trained on project specific
confidentiality procedures and required to sign an affidavit of non-disclosure and a
confidentiality agreement (see Appendices B and C).
2.2
Telephone Survey Quality Control Procedures
RTI project staff implemented several quality control procedures during the telephone
data collection period to ensure that data of the highest possible quality would be collected. The
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
4
Final Report
December 23, 2005
first effort to ensure that we collected high quality data was made during the interviewer training
sessions. The trainer and telephone supervisor observed each interviewer's performance during
the session, and made sure that extra attention and help were given to those interviewers who
had difficulty with a procedure or question specification. Interviewers were not allowed to begin
work on the project unless they had performed satisfactorily during the training. In addition, all
monitors, team leaders, supervisors and interviewers were required to take an oral exit exam at
the end of training, and those who did not pass were required to spend more time learning the
procedure in question (that they failed to answer correctly on the exit exam) and more time
practicing mock interviews before they were allowed to begin work on the survey.
The second quality control measure implemented during the telephone data collection
period involved silent monitoring of each interviewer's work. Telephone supervisors and other
project staff used RTI's computerized silent monitoring system to unobtrusively listen to a
sample of calls made by all telephone interviewers. Interviewers were provided both written and
verbal feedback on their performance to use to improve their performance for the remainder of
the study. Monitoring was conducted throughout the entire data collection period.
As a third quality control measure, project staff held regular meetings with the telephone
interviewers and supervisors throughout the data collection period. During these meetings,
multiple issues were discussed including operating issues such as production, the results of
quality control monitoring, and problems/issues that the interviewers noticed that could
potentially impact the response rate. As a result of the information gleaned through these open
discussions with the interviewers, project staff prepared and provided interviewers with scripts
containing additional statements to clarify the purpose of the survey for the respondent, why it is
important, and gave the interviewers instructions on how to handle problematic situations as they
arose.
Finally, RTI project staff actively monitored both production levels and the distribution
of cases across event and status codes. We monitored the number of refusals and the status of
refusal conversion and tracing cases. Each case designated by our call center as needing tracing
was reviewed to make sure that all roster lines had been attempted and efforts exhausted on those
roster lines, before sending the case to our tracing unit for intensive tracing. These steps allowed
us to identify problems early on and ensure that the sample was being worked efficiently.
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
5
Final Report
December 23, 2005
2.3
Tracing Results
The process of tracing for the MilCohort Non-Response Study was initiated for all
sample members up front during batch tracing as well as throughout data collection if the contact
data proved insufficient for contacting a sample member by telephone. Tracing was critically
important for locating individuals who may have moved or changed telephone numbers since the
last round of the Millennium Cohort Study data collection or for individuals who did not have a
working telephone number prior to the start of the MilCohort Non-Response Study. RTI’s
tracing efforts yielded both address and telephone number updates, however only updated
telephone number were used for this telephone non-response study.
Tracing was conducted in two steps. First, available information from sample lists was
submitted to vendors who matched this information with publicly available databases; this is
referred to as “batch tracing”. Second, intensive tracing was applied to sample members for
which calls had been attempted but the sample member was not located and for which a
telephone number was lacking after batch tracing. In intensive tracing, trained tracing staff
would attempt to locate a sample member through a series of intensive locating steps. We
describe each of these tracing steps in further detail below.
2.3.1
Batch Tracing
The primary objective of conducting batch tracing was to provide updated sample
contacting information in an efficient manner. Three ways we achieved the primary objective
included: 1) decreasing the number of cases that required individual, intensive tracing,
2) decreasing the cost of intensive tracing by providing tracing staff with the maximum amount
of contacting information available for subsequent locating steps, and 3) maximizing the number
of cases that could be interviewed over the telephone. After obtaining the sample, we sent the
sample file to an outside vendor to perform three batch tracing steps: National Change of
Address (NCOA), Telematch, and COMSERV, Inc.
•
NCOA is a database consisting of change of address data submitted to the U.S. Postal
Service.
•
Telematch is a computerized telephone number service that uses a name, street
address and zip code to match against residential white pages to develop a telephone
number.
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
6
Final Report
December 23, 2005
•
COMSERV is a computerized system containing a comprehensive database of
information on deceased individuals. COMSERV pioneered the unique DIS concept
in 1980 with a database of approximately 23 million individuals.
Batches of contacting information were submitted first to NCOA to obtain an updated
address (if the sample member had filed a change-of-address with the U.S. Postal Service). The
CATI database was updated with any new information, and then records were submitted to
Telematch for a telephone number update. New information obtained through batch tracing was
appended to records for each sample member without overwriting existing information.
The batch tracing results from NCOA and Telematch provided an early indication that
tracing the sample was going to be a challenge. From NCOA, we received a match rate of 9
percent and a corrected zip code rate of 33 percent. The NCOA match rate indicates the number
of names and addresses that match to their database. The normal NCOA match rate is between 2
and 8 percent. Files with match rates that fall outside of the expected range often indicate
significant address problems. For Telematch, we received updated phone numbers for
approximately 40 percent of the sample. This match rate is significantly less than other telephone
studies at RTI which typically receive a match rate between 45 percent and 55 percent. For
COMSERV, we received five records confirming a deceased sample member. COMSERV
matched on social security number (SSN,) date of birth, and first and last name. Exhibit 2-2
displays a summary of our batch tracing results.
Exhibit 2-2. Batch Tracing Results
Vendor
NCOA
Telematch
COMSERV, Inc
Number of Updates Received
274
1,193
5
Match Rate
9%
40%
0.2%
After batch tracing, we had approximately 485 cases for which we had either a phone
number with a non-existent or missing area code or were missing a phone number entirely. We
put these cases on hold until we could determine if we had adequate tracing resources to conduct
intensive tracing on these cases.
2.3.2
Intensive Tracing
The second component of tracing was intensive tracing. Cases were routed to intensive
tracing if a call (or several calls, as necessary) had been attempted by an interviewer but the
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
7
Final Report
December 23, 2005
potential respondent was not located or if cases lacked a telephone number after batch updates.
Exhibit 2-3 displays a summary of the individual submissions we sent to intensive tracing.
Exhibit 2-3. Summary of Individual Submissions to Intensive Tracing
Batch
Number
of Cases
Date Submitted to
Intensive Tracing
Level of
Effort
1
200
Cases containing phone numbers
that proved to be not viable after
work in the Call Center
September 19th
15 minutes
500
Cases containing phone numbers
that proved to be not viable after
work in the Call Center
2
September 28th
15 minutes
3
350
Cases containing phone numbers
that proved to be not viable after
work in the Call Center
October 5th
30 minutes
4
50
Cases containing a bad area code
or no phone number
October 5th
15 minutes
433
Cases containing a bad area code
or no phone number
5
Type of Case
Locate
Rate
73%
72%
65%
th
October 19
15 minutes
During the intensive effort, tracing staff reviewed the contacting information associated
with a particular sample member. The tracer would then perform a series of real-time searches of
the contacting information available in consumer credit bureau databases that might help locate
the particular sample member. Additionally, tracers would search a series of proprietary
databases for contacting information that included information culled from public and financial
records. Other sources that tracers used included Directory Assistance, reverse directories,
residential sources, specialized directories, and other consumer databases.
After intensive tracing yielded viable contacting information for a sample member, a
tracer would confirm that the number was a working telephone number before submitting located
cases back to the call center. On average, each tracer worked a case for a 15 minute level of
effort. This initial 15 minute level of effort yielded an overall 73 percent locate rate for the first
700 cases. Approximately half way through intensive tracing, we conducted an experiment with
350 cases to see if an additional 15 minute level of effort would yield a higher locate rate. We
found that the additional 15 minutes did not increase the overall locate rate (72 percent).
We followed a similar multi-step process for cases which had bad contact information at
the beginning of the study once it became clear that we had sufficient resources to conduct
intensive tracing for these cases. Initially, we sent 50 cases with bad contact information to
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
8
Final Report
December 23, 2005
intensive tracing with a 15 minute level of effort. These efforts resulted in a 68 percent locate
rate. After evaluating our resources again, we determined that we had sufficient resources to
send the remaining 433 bad contact cases to intensive tracing. The overall locate rate for the bad
phone number cases was 65 percent which is approximately 8 percent less than we achieved for
cases where we initially had usable contact information.
Overall, of the 1,533 sample members who required intensive tracing, viable contacting
information was identified for approximately 70 percent of the cases (n=1,079). Intensive tracing
activities began on September 19th and ended October 27th. Exhibit 2-4 summarizes the outcome
of the intensive tracing activities.
Exhibit 2-4. Intensive Tracing Results
Summary of Cases
Number
Located
Unable to Locate
Total Number of Cases Sent to Intensive Tracing
2.4
%
1,079
70%
454
30%
1,533
100%
Call Outcomes
Data collection began on September 8, 2005, and ended on November 9, 2005. If a
telephone number was provided, interviewers attempted to contact the sample member. Any
cases with incomplete contacting information were submitted for tracing, and subsequently
routed to telephone interviewing once new contacting information was obtained. Originally, RTI
budgeted an average number of 4 attempts for all cases. Due to this budget constraint, we
targeted calls to households during times that have been shown to be most productive in terms of
contacting households and completing interviews. For the first two call attempts, we called
during the evening/weekend time slots to maximize our chances of reaching sample members at
home.
As data collection progressed and as we reviewed our budget carefully, we increased the
number of call attempts in order to increase our chances of completing additional interviews with
the harder to reach sample members. Exhibit 2-5 displays the dates and time we increased the
maximum number of attempts throughout the data collection period. A majority of the increases
took place prior to the weekend when we experienced higher contact rates.
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
9
Final Report
December 23, 2005
Exhibit 2-5. Increase in the Number of Call Attempts by Date
Number of Call Attempts
8 attempts
12 attempts
15 attempts
18 attempts
Day of the Week Call Attempt
Increase Occurred
Friday
Friday
Wednesday
Friday
Date Call Attempt Increase
Occurred
September 23rd
October 7th
October 12th
October 21st
During data collection, we tracked the mean call count by summary status group. The
mean number of call attempts by summary status group is displayed below in Exhibit 2-6. The
average number of call attempts for the ineligible cases was 6, for the not located cases was 4,
for the not interviewed cases was 9, and for the completed cases1 was 5. The overall mean call
count for the entire sample was 7. As anticipated, the harder to reach non-interview cases took
substantially more effort to follow-up than did the completed cases. In addition, even the
completed cases took more than the originally budgeted 4 attempts.
Exhibit 2-6. Mean Call Count by Summary Status Group
Final Status
Ineligible
Not Located
Not Interviewed
Completed
Overall
Mean Call Count
6
4
9
5
7
During data collection, we also tracked the distribution of attempts for the completed
cases. In Exhibit 2-7, we show that approximately 54 percent of the completed interviews
(n=484) took place during the first 4 attempts. However, to increase our yield of completed
interviews, we had to increase our number of attempts. The number of completes shows a steady
decrease the higher the attempts with the exception of a slight increase between attempts 6 and 7
and a more significant increase between attempts 8 and 9.
1
Of 16 NHRC completed cases, 4 were contacted but did not complete the non-response interview, 7 were not
contacted, 5 were not called.
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
10
Final Report
December 23, 2005
Percent
Exhibit 2-7. Percent of Completed Cases by Call Attempt
18
16
14
12
10
8
6
4
2
0
1
2
3
4
5
6
7
8
9 10 11 12 13 14 15 16 17 18 19
Number of Call Attempts
By raising the maximum call attempts to 8, we obtained an additional 228 completed
interviews (25 percent), raising the maximum call attempts to 12 resulted in an additional 148
completed interviews (17 percent), raising the maximum call attempts to 15 resulted in an
additional 26 completed interviews (2 percent), and raising the maximum call attempts to 182
resulted in an additional 14 completed interviews (2 percent).
2.5
Data Collection Outcomes
Data collection activities resulted in a response rate of 31 percent. The response rate for
the MilCohort Non-Response Study was calculated by taking the number of completed
interviews divided by the number of interviews plus the number of non-interviews plus cases of
unknown eligibility. Its value is equal to the product of the contact and cooperation rates, which
for this study was equivalent to AAPOR Response Rate 4 (RR4)3. Exhibit 2-8 shows the final
disposition of cases which was used in the response rate calculation.
Refusal cases were isolated in a different queue within the call scheduler so that only our
most experienced interviewers were able to work those cases. Just under 6 percent (n=172) of
sample members ever refused during the non-response study4. Refusals from sample members
2
Three cases received 19 attempts. The maximum call attempts setting can be overridden if there has been an
appointment set immediately prior to max call attempt.
3
American Association of Public Opinion Research, 2000.
4
172 sample members initially refused the non-response study and received a refusal strength rating of mild, firm, or
hostile. An additional 58 cases were coded as refusal by other and were not given a refusal strength rating since
these were not sample member refusals.
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
11
Final Report
December 23, 2005
occurred for a variety of reasons that were categorized by reason and strength of the refusal
within the CATI system. Interviewers were able to code multiple reasons for refusal per case.
The most cited reason for refusal was being not interested in the study (n=97, 43 percent),
followed by some other reason (n=62, 27 percent), being too busy or not having time to
participate in the survey (n=27, 12 percent), hanging up before or during the introduction (n=24,
11 percent), and 2 percent or less for each of the following 3 reasons: asked about the purpose of
the study (n=4, 2 percent), asked how long the survey would take (n=3, 1 percent), and asked
how his/her name was selected (n=5, 2 percent).
Exhibit 2-8. Final Dispositions
Number in Sample
Completed Qx/Interviews
Completed Non-Response Interview
Total
3,000
No.
Percent
908
908
30.3%
30.3%
5
41
16
62
0.2%
1.4%
0.5%
2.1%
Other Final Non-Interviews
Final Subject Located/Verified
Final Subject Not Yet Located/Verified
Final Refusal
Final Other non-interview
Final Out of Country
Final Unable to Locate
Final Language Barrier
Partial Complete
Total Other Final Non-Interviews
27
950
117
82
98
704
25
27
2,030
0.9%
31.7%
3.9%
2.7%
3.3%
23.5%
0.8%
0.9%
67.7%
Total Final Cases
Total Eligible Cases
3,000
2,938
100.0%
Ineligible Cases
Ineligible - Deceased
Not ever released - due to Hurricane Katrina
NHRC reclassified as MilCohort responders
Subtotal Ineligible Cases
Raw Response Rate
Adjusted Response Rate
30.3%
30.9%
Just over one-half of the refusals received were mild refusals (n=88), that is the sample
member conveyed his or her concern in a non-threatening manner. An additional 44 percent
(n=75) were firm refusals and 5 percent (n=9) were hostile refusals. Interviewers attempted up to
two contacts with the sample member to convince him or her to participate in the MilCohort
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
12
Final Report
December 23, 2005
Non-Response Study after the initial refusal. Hostile refusals, including threats of legal action
and other statements that indicated to the interviewer or project staff that the sample member did
not want to participate in the non-response study, were not contacted a second time; rather they
were coded as final refusals after the first occurrence.
Ineligible cases consisted of cases where we determined the sample member was
deceased, could not be contacted because their phone number fell into the regions devastated by
Hurricane Katrina, or they completed the full NHRC interview during the non-response study
field period. Only 2 percent of cases (n=62) were determined to be ineligible.
Other final non-interview cases consisted of final subject located/verified, final subject
not yet located/verified, final refusals, final other non-interviews, final out of the country cases,
final unable to locate cases, final language barrier cases, and partial interviews. Final subject
located/verified cases are cases where we spoke to the sample member but were unable to
complete an interview during the non-response survey period. Final subject not yet
located/verified cases were cases where we had a viable telephone number, but were unable to
speak directly to the sample member during the non-response survey period, whereas final
unable to locate cases were cases that we determined the telephone number was not viable and
we were not able to locate these sample members even with intensive tracing efforts. The final
unable to locate cases include cases that were deemed unlocatable in our tracing unit and in our
telephone unit. The other final non-interview cases included those cases where the sample
member was unavailable for the duration of the study, incapable, or institutionalized.
2.6
Data Collection Debriefing Summary
On November 10th, we held debriefing meetings with our tracers and interviewers. The
purpose of the debriefing was to determine what went well and what could be improved for a
potential next round of data collection. Prior to the meeting, each tracer and interviewer received
a debriefing form (see Appendix D). They spent approximately 15 minutes filling out the form
and then brought their completed forms to the debriefing meeting. Below we list the highlights
from the tracing and interviewing debriefing meetings.
2.6.1
Highlights from the Tracing Debriefing
Overall, the tracers enjoyed working on this study. The population was easier to trace
than some of their more difficult studies where they are trying to find people with minimal
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
13
Final Report
December 23, 2005
locating information. They all remarked on the luxury of having an SSN to work with during an
intensive tracing.
•
Tracers found that the most useful sources for intensive tracing proved to be Experian
and Fastdata, which are the two interactive databases that contain credit header
information.
•
Several tracers mentioned that cell phones were a problem on this study. We cannot
trace cell phones exclusively. If a cell phone is associated with a person's name, we
will get that information through normal search engines. We can also check a listing
to see if a phone number is a cell phone, but this listing will not associate the cell
phone with an individual.
•
In terms of e-mail addresses, sometimes the tracers collected this from contacts. They
do have the ability to search for e-mails through multi-people search, but it is only
useful if we have a unique name. Unfortunately, this search does not allow you to
match on SSN. It would be very time consuming to do this search.
•
The tracers said they really did not notice a difference between the bad phone number
cases and the good phone number cases.
2.6.2
Highlights from the Interviewer Debriefing
Overall, the interviewers really enjoyed working on this study. They repeatedly remarked
on the fact that this population overall was very cooperative if they could get them on the phone.
•
Interviewers remarked that the six hour project training was adequate in preparing
them to gain cooperation and administer the interview. They found the study
information on the MilCohort Study website very helpful when answering respondent
questions.
•
Many interviewers remarked that they found the cases became easier to work as the
data collection period progressed due to the located cases they received from the
intensive tracing unit. Many interviewers remarked on their initial frustration at
encountering so many disconnected numbers early in the data collection process.
•
Several telephone interviewers commented on the difficulty of reaching active
military members when they were on base or on a ship. Several interviewers said that
having the person’s service number would be helpful when dealing with switchboard
operators at these facilities.
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
14
Final Report
December 23, 2005
•
Overall, the interviewers commented that speaking with military staff was easier
since they seemed more willing to participate than most other study sample members.
It took more effort to convince the no longer active military staff to participate since a
majority of them felt that they no longer had to do the study now that they were out of
the service.
•
When asked about the Blaise instrument, interviewers commented that the survey was
easy to administer. The only suggestion they had for improvement involved changing
the flow of the introduction so that they identify their names, the Department of
Defense, and then RTI. Many interviewers remarked on sample member’s
willingness to participate when they heard we were calling for the Department of
Defense.
•
Many interviewers commented on the helpfulness of having the date and year the
sample member completed the last survey. They repeatedly used this information to
remind sample members about the study. One interviewer commented that a
respondent had told her it would be helpful if they marketed the survey at the unit
level.
•
Several interviewers received recent complaints from sample members regarding the
amount of mail they had received and the fact that they had received it via certified
mail. The interviewers felt that in these instances, it was very difficult to gain
cooperation from these sample members who were angry about receiving so much
mail and having to go to the post office to sign for the survey.
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
15
Final Report
December 23, 2005
3. Questionnaire Results and Recommendations
In this chapter we present the questionnaire results, as well as recommendations for
future waves of the MilCohort Study and for potential future non-response studies. The first
section presents the results and initial recommendations, and the second section presents a
summary of recommendations. A frequency listing with the survey results appears in
Appendix E.
3.1
Results and Initial Recommendations
Reasons for Non-Participation. Sample members were asked for the two main reasons
they did not participate in the current round of the Millennium Cohort Study. The question was
open-ended with pre-set answer choices that interviewers used to code the respondents answer. If
the answer did not clearly fit into one of the pre-set answer choices, the interviewer was
instructed to code the answer as “other” and enter the text for that response. Exhibit 3-1
illustrates the main reason respondents cited for not participating in the current round of the
MilCohort Study.
Exhibit 3-1. Main Reason for not Participating in this Round of the Study
n=258
30
n=217
Percent
25
20
n=158
15
10
5
n=65
n=44
n=86
n=51
n=21
n=5
0
Not recall No w eb Deployed Survey
invitation access
too long
Out of
military
now
Health
Not
Other
interested
No 2nd
answ er
First Reason for Non-Participation
Currently being out of the military (28 percent) and not recall getting an invitation (17
percent) were the two most chosen answers for not participating, followed by not being
interested (9 percent), being deployed (7 percent), survey was too long (6 percent), could not
access the web (5 percent), and health kept sample member from completing (1 percent).
Additionally, 24 percent of respondents gave some other reason for not participating. Of the 217
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
16
Final Report
December 23, 2005
respondents who gave an “other” answer, the main reasons cited were: too busy (29 percent),
thought they had already completed the survey (13 percent), didn’t recall getting an invitation (9
percent), personal reasons (6 percent), just didn’t do it (6 percent), sample member had moved (6
percent), while the remaining 26 percent of answers were too varied to be summarized.
Exhibit 3-2 illustrates the second reason cited for not participating in the study. The two
second reasons mentioned most often by respondents were not being interested (9 percent) and
out of the military (9 percent), followed by not recall getting an invitation (7 percent), survey
was too long (5 percent), could not access the web (4 percent), being deployed (2 percent), and
health kept sample member from completing (1 percent). “Other” responses were given by 161
respondents (44 percent) to the second reason for not participating in the study. These responses
can be summarized as: too busy (42 percent), lost the paper copy (11 percent), thought they had
already completed the survey (7 percent), didn’t recall getting an invitation (7 percent), sample
member had moved (6 percent), just didn’t do it (3 percent), and the remaining 23 percent of
“other” were too diverse to summarize.
Percent
Exhibit 3-2. Second Reason for not Participating in this Round of the Study
50
45
40
35
30
25
20
15
10
5
0
n=331
n=160
n=66
n=65
n=53
n=38
n=12
n=30
n=6
Not recall No w eb Deployed Survey Out of
invitation access
too long Military
Health
Not
Other
interested
No 2nd
answ er
Second Reason for Non-Participation
Recommendations. Based on the answers to this question, we have 6 recommendations
for consideration. First, we recommend that the study materials be revised to emphasize that the
participation of sample members who are not in the military is just as important as those who are
still in the military. Next, we would recommend tailoring some study materials (e.g., lead letters)
for those sample members who are not currently in the military. Third, we recommend revisiting
the panel maintenance procedures to see if improvements can be made in maintenance of contact
information for sample members. Fourth, we recommend continuing to batch trace all sample
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
17
Final Report
December 23, 2005
members before each round of survey implementation. This helps ensure that a maximum
number of sample members receive the survey request. Fifth, a majority of the “other” responses
cited being too busy to participate in the study. We recommend emphasizing that respondents
can stop and start the interview on the web at their convenience as a way to minimize the impact
on their schedules for completing the survey. Finally, since nearly 15 percent of respondents to
the “other” category mentioned thinking they had already completed the survey, we recommend
emphasizing that the study is longitudinal and that NHRC will be asking for their participation in
a similar survey every 3 years.
Awareness of Incentive Offer. Exhibit 3-3 shows that nearly two-thirds (64 percent) of
respondents stated they were not aware of the incentive offer. The survey research literature
consistently demonstrates that incentives are a cost effective way of obtaining participation. 5
Exhibit 3-3. Awareness of MilCohort Incentive Offer
70
n=579
60
Percent
50
40
n=322
30
20
10
0
Yes
No
Aware of Incentive?
Recommendations. Based on the report that respondents are not aware of the incentive
offer, we recommend increasing the visibility of incentive offer in the study materials and on the
study website. This increased visibility would need to be balanced with other survey goals, and
careful consideration of how to implement this recommendation so that the interaction with
sample members remains viewed as a social exchange rather than an economic exchange. One
theory of survey participation posits that keeping a survey request as a social exchange invokes
social norms that if a request is made by someone it is complied with if it doesn’t take too much
effort. Offering an incentive can be viewed as an economic exchange, however, if the incentive
5
Dillman, D.A. 2000. Mail and Internet Surveys: The Tailored Design Method. New York: Wiley.
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
18
Final Report
December 23, 2005
is not introduced as a token of appreciation (which acknowledges that the survey team cannot
adequately compensate the respondent for their time) and if the amount is large enough to make
it viewed as a payment rather than a token of appreciation.
Items to Encourage Participation. Respondents were reminded of the current incentive
offers, and then asked to choose their top two choices among a list of six items that would better
encourage participation. To distinguish between types of responses indicating no incentive was
necessary, two other answer options were included but not read to the respondent. These
included: nothing because a gift is not necessary and nothing because the respondent would not
complete the survey even with receipt of a gift for participation. See Exhibit 3-4 for illustration
of results.
Percent
Exhibit 3-4. First Item Chosen to Encourage Participation
35
30
25
20
15
10
5
0
n=263
n=239
n=102
n=84
n=45
n=38
n=57
n=41
n=22
G if
Ca
Ot
Pa
Co
Mi
Clo
No
No
he
tc
id
sh
len
ffe
thi
thin
r
ar d thi ng
su
ni u
em
ng
g
b
s
-w
s
m
w
u
c
g
ou
/
g
r
C
s
i
i
f
s
p
t
er v
oh
sn
ld n
ti o
o rt
i ce
nt
ot
ot
om
Co
ne
sp
do
c
in
e
a
es
su
c
g
ifi c
azi
sar
rv e
l og
ne
y
y
o
First Item to Encourage Participation
Among the item first mentioned, two choices received similar endorsement from
respondents: gift cards (29 percent) and cash (26 percent). Receipt of clothing with a service
specific logo was a distant third choice (11 percent), followed by a paid subscription to a
magazine (9 percent), coffee mugs (6 percent), would not do survey even with gift (5 percent),
Millennium Cohort coin (5 percent), no gift necessary (4 percent), and some other answer (2
percent). Nearly 37 percent of “other” responses mentioned choices in the list read to
respondents: 5 percent cited gift cards and 32 percent said phone cards. The remaining 63
percent can be categorized as: choose an incentive (9 percent), shorter survey (9 percent), other
physical items (9 percent), nothing (5 percent), results from the survey (5 percent), and 27
percent were too varied to be summarized.
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
19
Final Report
December 23, 2005
Exhibit 3-5 provides the results to the second item chosen by respondents to encourage
participation. The second items chosen by respondents did not differ significantly in order
compared to the first items chosen, however the distribution among response options changed
somewhat. Cash was the most requested item (22 percent) followed by gift cards (19 percent),
clothing with a service specific logo (16 percent), paid subscription to a magazine (11 percent),
coffee mug (9 percent), no gift necessary (8 percent), would not do survey even with gift (6
percent), Millennium Cohort coin (5 percent), and some other answer (5 percent). Nearly 78
percent of “other” responses mentioned choices in the list read to respondents: 56 percent cited
phone cards, 14 percent cited gift cards, and 8 percent said t-shirt. The remaining 22 percent can
be categorized as: 11 percent were other physical items, 3 percent wanted results from the
survey, and the remaining 8 percent were too varied to summarize.
Exhibit 3-5. Second Item Chosen to Encourage Participation
25
Percent
n=165
n=139
20
n=118
15
10
5
n=57
n=67
n=42
n=82
n=38
n=36
0
G if
Ca
Ot
Co
Clo
No
Mi
Pa
No
he
tc
sh
len
i
thi
ffe
thi
thin
r
ard
ng
ni u
e m d s ub
ng
g
s
-g
m
-w
w
s
u
c ri
/s e
gs
C
ifts
ou
pti
oh
rvi
ld
no
on
o rt
ce
no
tn
to
Co
sp
td
e
m
c
in
ec
os
es
a
g
ifi c
sar
urv
a
zi n
l og
y
ey
e
o
Second Item to Encourage Participation
Recommendations. We have two main recommendations for utilizing the answers given
by respondents on ways to better encourage participation in the MilCohort Study. First, we
recommend serious consideration be given to whether to add gift cards and/or cash to the current
incentive offer. We recognize the limitation that cash cannot be used in government/military
surveys. However, given the importance of this longitudinal study and the evidence that nonrespondents (who may differ in important ways from respondents) indicated this would
encourage their participation may warrant a re-visitation of this issue. Second, the choice of
clothing with a service specific logo was a distant third among the choices allowed, 220
respondents (or 24 percent) did cite this as an item to encourage participation. The question
asked only mentioned clothing with a service specific logo rather than also mentioning clothing
with a MilCohort logo. Due to this limitation, we cannot clearly say whether clothing with a
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
20
Final Report
December 23, 2005
service specific logo is preferred by MilCohort non-response study participants. In a future nonresponse study, we recommend consideration be given to asking specifically whether MilCohort
participants would prefer a MilCohort or service specific logo so the MilCohort team can more
clearly evaluate the preference of MilCohort participants.
Magazine Specified. If respondents indicated that a paid subscription to a magazine was
one of their top two choices for encouraging participation in the survey, they were then asked to
name a magazine that would accomplish this goal. Exhibit 3-6 presents the results from this
question.
Exhibit 3-6. Magazines Cited as Way to Encourage Participation
35
30
n=118
Percent
25
n=31
20
n=21
15
n=18
10
n=18
n=9
5
n=13
n=6
n=2
0
Sp
o rt
Mi
Wo
Ne
li ta
ws
me
ry
s /O
/Po
n's
l
u td
/M
i tic
en
s
oo
's
r
He
al t
h
Sc
Ot
Co
Do
he
ien
mp
n't
r
ce
kn
u te
ow
r
Type of Magazine Specified
The magazines cited as a way to encourage participation fell into 9 general categories:
sports/outdoors (28 percent), women’s/men’s (19 percent), news/politics (13 percent), military
(11 percent), don’t know (11 percent), other (8 percent), health (5 percent), science (4 percent),
and computer (1 percent).
Recommendation. One factor to consider when deciding whether to add magazines to the
incentive options is the administrative cost of providing this option. The administrative costs
alone may prove prohibitive. Further, due to the diversity of magazines cited, as well as the low
percentage of respondents who chose a paid subscription to a magazine as a better way to
encourage participation (9 percent as first choice, 11 percent as second choice), we do not
recommend using magazines as an incentive choice.
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
21
Final Report
December 23, 2005
Amount of Cash Specified. Respondents who indicated that cash was a way to better
encourage survey participation were asked to specify an amount that would achieve this goal.
See Exhibit 3-7 for a summary of responses. Encouragingly, 63 percent of respondents indicated
that $25 or less would encourage participation (25 percent chose $10 or less; 37 percent chose
$11 - $25), with only 22 percent indicating $26-$50, 8 percent stating $51 to $100, and only 7
percent giving a value over $100.
Exhibit 3-7. Amount of Cash Suggested to Encourage Participation
40
n=132
35
Percent
30
25
n=90
n=78
20
15
n=30
10
n=25
5
0
$10 or Less
$11-$25
$26- $50
$51 - $100
More than
$100
Cash Amount Specified
Recommendations. We have two recommendations for the amount of cash to offer, if
cash is deemed an appropriate incentive option for respondents who are no longer in the military.
(We recognize that cash incentives are not permissible for active duty respondents.) First, we
recommend that the MilCohort Study team conduct an experiment to determine whether
participation rates (and other key study indicators) differ by receipt of a $10 versus $20 cash
incentive offer. Consideration should be given to conducting this experiment with a small
segment of the entire study population so that the most useful and most cost effective incentive
amount could be chosen for future rounds of the survey. Second, a $20 incentive could be used
right away to increase participation. This amount is in line with the amount of cash specified by
respondents to this study.
Health Rating. Prior to fielding of the non-response study, it was hypothesized that
some segment of the population may not be participating due to poor health. Given that the main
point of the Millennium Cohort Study is to track the long term health of military members, it was
deemed important to get a health indicator during this initial non-response study. The health
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
22
Final Report
December 23, 2005
rating question used in the MilCohort survey was also used here, and asked respondents to rate
their health excellent, very good, good, fair, or poor. Exhibit 3-8 shows the responses to this
question.
Exhibit 3-8. Health Status Reported by MilCohort Non-Response Study Participants
40
n=309
35
n=264
Percent
30
25
n=225
20
15
n=87
10
5
n=19
0
Excellent
Very good
Good
Fair
Poor
Health Status
A majority of respondents (88 percent) reported their health status as either excellent (25
percent), very good (34 percent), or good (29 percent). Only 10 percent of non-response
participants rated their health as fair, and 2 percent as poor.
Recommendation: Since we did not directly ask about the impact of health status on
participation in the MilCohort Study, we have no recommendation about altering survey
procedures or study materials.
Survey Notification. A key component of obtaining survey cooperation is in ensuring
that the survey request is received by sample members. The number of non-response cases that
had to undergo tracing (see discussion of tracing results, section 2.3, above for more details)
provides some evidence that MilCohort sample members may not be receiving the survey
request. Respondents were explicitly asked whether they received any mail or e-mail
notifications about the study. While more respondents reported receiving mail notifications of
the study than e-mail, Exhibit 3-9 shows that only 39 percent of respondents said they received
any mail notification and only 15 percent received e-mail notification. To help bring this finding
into context, we compared the answers for those respondents who indicated on Question 1 that
one of the two main reasons they did not participate in the last round of MilCohort was because
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
23
Final Report
December 23, 2005
they did not recall receiving an invitation to their answers on receipt of mail (Question 7) or email (Question 8) survey notification. Of the 211 respondents who indicated that they did not
recall getting an invitation on Question 1, only 17 (8 percent) indicated they received either a
mail survey notification on Question 7 (n=15) or received an e-mail notification on Question 8
(n=1), or said yes to both Questions 7 and 8 (n=1).
A note of caution is needed about our survey notification and tracing results: while we
obtained updates for both telephone and address information as a result of our tracing efforts, all
of our contacts with sample members were conducted via telephone. Therefore, we cannot
clearly state evaluate the accuracy of currently available address or e-mail address information
since we did not attempt to contact sample members via those modes for the non-response study.
Exhibit 3-9. Percent of Non-Response Study Participants
Who Indicated They Received Mail or E-mail Survey Notifications
45
40
n=357
35
Percent
30
25
20
n=134
15
10
5
0
Mail
Email
Type of Notification Received
Recommendations. We have three recommendations for updating survey notifications.
First, we recommend continuing panel maintenance activities to ensure that the most up-to-date
contact information is available at the start of each survey wave. This should include continuing
to send an e-mail to the e-mail address of sample members who participated via the web on the
prior round. Next, as discussed above, we recommend continuing routine batch tracing of sample
members prior to beginning each data collection wave. Finally, we recommend continuing a
second round of batch tracing mid-way through the survey period for all non-respondents. This
would allow the updating of any new contact information and enable the survey team to
maximize participation rates.
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
24
Final Report
December 23, 2005
Internet Access. Exhibit 3-10 presents respondents answers about internet access. When
asked whether they had internet access to complete the survey, 83 percent indicated they did
have internet access. This finding is encouraging about the potential usefulness of the web for
MilCohort. Allowing sample members to complete the interview over the web is a cost effective
means of data collection, and can be convenient for respondents and potentially reduce the
burden of survey participation.
Exhibit 3-10. Internet Access of MilCohort Non-Response Study Participants
90
n=758
80
70
Percent
60
50
40
30
n=147
20
10
0
Yes
No
Have Internet Access?
Recommendation. We recommend that all study materials, including those being mailed
to sample members and the web site, emphasize the availability of the web for survey
participation. This can be done both through text and through the visual display of the
information.
Requested Survey Notification. Participants were asked whether they would like to
receive another copy of the survey so they could participate in this round of the MilCohort
Study. Only those participants who indicated they had e-mail access were asked if they would
like to receive another e-mail notification about the study, but all participants were asked if they
would like to receive another paper version of the survey. Exhibit 3-11 shows that 76 percent of
respondents who indicated they had an internet access requested the online survey link to
complete the survey, and 34 percent of all respondents asked to receive another hard copy
survey. These results are encouraging because they show that non-respondents to the current
MilCohort wave are reconsidering the survey request after participation in this non-response
study. By linking completed interviews to the study id for non-response cases, the MilCohort
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
25
Final Report
December 23, 2005
Study team will be able to determine whether the non-response study participants eventually
completed the interview.
Exhibit 3-11. Survey Notification Requests
80
n=573
70
Percent
60
50
40
n=308
30
20
10
0
Via email
Hard copy
Type of Survey Request
Additionally, of the 573 respondents who provided an e-mail address for NHRC to send
another version of the survey request, 345 (60 percent) were different from the sample file e-mail
address provided by NHRC. Of the 308 respondents who indicated they would like NHRC to
send them another paper copy of the questionnaire, 217 cases provided updated contact
information. Of those 217 cases:
•
87 provided first name, 62 provided middle name, 87 provided last name. Of these
cases, 10 names were significantly different from the sample file name.
•
90 provided a 1st line address, 14 provided a 2nd line address. Of these cases, 78
addresses were significantly different from the sample file address.
•
90 cases provided city. Of these cases, 30 were different from the sample file city.
•
90 cases provided state. Of these cases, 22 were different from the sample file state.
•
90 cases provided zip code. Of these cases, 67 were different from the sample file zip
code.
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
26
Final Report
December 23, 2005
Recommendations. First, if participation in the non-response study increases participation
in the MilCohort then we would advise that telephone prompting be considered as an addition to
the MilCohort data collection protocol. Next, we would also advise that future non-response
studies be completed periodically so the MilCohort team can continue to assess reasons for nonresponse and address those issues as best they can before longitudinal participation in MilCohort
becomes a threat to the validity of the study.
Updated Contact Information. Participants who received an e-mail link to the survey
were asked to provide their e-mail address. Approximately 573 participants did provide an email address. Similarly, participants who requested a paper version of the survey were asked to
confirm their contact information. Exhibit 3-12 shows that 29 percent of the 308 participants
who asked for a hard copy of the survey provided updated contact information. This is another
indicator that the contact information recorded by the MilChort team may need updating more
frequently and prior to survey waves.
Exhibit 3-12. Updated Contact Information Provided by Participants
Requesting Hard Copy Questionnaire
80
70
n=217
Percent
60
50
40
n=90
30
20
10
0
No
Yes
Provided Updated Contact Information?
Additionally, of the 908 completed non-response interviews, 375 (41 percent) had
updated contacted information at the time of completion. Of these 375 cases with updated
contact information, 5 (1 percent) had updated address information only, 167 (45 percent) had an
updated phone number, and 203 (54 percent) had both an updated address information and phone
number.
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
27
Final Report
December 23, 2005
Recommendations. Approximately 60 percent of reported e-mail addresses differed from
the current e-mail information recorded for the sample members, which supports our
recommendations for implementing telephone prompting and confirmation of contact
information during the prompting call. The prompting call could also be used to offer to re-send
the survey request either via e-mail or hard copy.
3.2
Summary of Recommendations
The MilCohort Non-Response Study has provided very valuable information on reasons
for non-participation, as well as items that may encourage participation. Our overall
recommendations cover 6 main areas: study materials, panel maintenance, tracing sample
members, incentives, telephone prompting, and future non-response studies. Our
recommendations are summarized below:
1.
Study Materials
A. Revise materials to emphasize the importance of participation for sample
members no longer in the military.
B. Tailor materials, as possible, for sample members no longer in the military. For
example, customize lead letter to these sample members to acknowledge they
are no longer in the military and the importance of their participation.
C. Increase visibility of incentive offer on study materials and website.
D. Continue to emphasize that the survey can be completed on the web, both
through use of text and visual display of information.
E. Emphasize that respondents can complete the interview on the web in several
sessions thereby fitting the task into their schedule.
F. Re-explain or emphasize the longitudinal nature of the study, and clarify that
respondents will be asked to participate in similar surveys every 3 years. Also
emphasize the importance of their participation in each round to ensure the
validity of the longitudinal results.
2.
Panel Maintenance
A. Revisit procedures to see if improvements can be made in receipt and
maintenance of contact information.
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
28
Final Report
December 23, 2005
B. Continue panel maintenance activities to update contact information to ensure
that as many sample members as possible receive survey request.
3.
Tracing Sample Members
A. Revisit current tracing procedures. Since 22 percent (n=208) of the 908
completed non-response interviews obtained updated email or address
information (which are the main methods of contact for MilCohort as opposed to
telephone which was used for the non-response study), it is worth revisiting
current tracing (and panel maintenance) procedures to determine if further
actions would be beneficial for MilCohort.
B. Continue routinely tracing sample members before each round of survey
implementation.
C. Continue second round of tracing mid-way through the survey period.
4.
Incentives
A. Consider adding gift cards or cash as option.
B. Conduct experiment to determine optimal value for cash, or start with $20
incentive.
C. Consider asking a more specific question related to respondent preference for
clothing with a MilCohort logo compared to a service specific logo (or picture)
during a future round of non-response follow-up.
5.
Telephone Prompting
A. If participation in the non-response study increases participation in the
MilCohort then we would advise that telephone prompting be considered as an
addition to the MilCohort data collection protocol. For example, an attempt will
need to be made to disentangle the outcome of any mailings that occurred as
part of the MilCohort protocol versus mailings that occurred because
participants in the non-response study requested study materials be sent to them
again.
B. Further analysis showed that 60 percent of reported e-mail addresses differed
from e-mail addresses in the sample file, and at least 41 percent (n=90) of the
cases requesting a paper copy of the survey provided at least one piece of
updated contact information. Receiving this number of updates from both
participation in the non-response study, and tracing efforts argues for
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
29
Final Report
December 23, 2005
considering implementation of telephone prompting and confirmation of contact
information during the prompting call.
C. Verify contact information during telephone prompting.
D. Offer to re-send survey request either via e-mail or hard copy during telephone
prompting.
6.
Future Non-Response Studies
A. Consider conducting a larger scale non-response study with all MilCohort Study
non-respondents.
B. We would also advise that future non-response studies be completed
periodically so the MilCohort team can continue to assess reasons for nonresponse and address those issues as best they can before longitudinal
participation in MilCohort becomes a threat to the validity of the study.
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
30
Final Report
December 23, 2005
Appendix A – Questionnaire
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
Final Report
December 23, 2005
Millennium Cohort Study: Pilot Study of 3,000 Non-Responders
Final Questionnaire - Version 3.7
CATI Specifications
August 30, 2005
Intro_1 - Introduction
Hello, I am calling on behalf of the U.S. Department of Defense, my name is ______________ and I
am from the Research Triangle Institute (RTI). I am trying to locate [RESPONDENT NAME] about
an important study he/she is participating in.
May I please speak with [RESPONDENT NAME]?
1 = YES, SUBJECT IS AVAILABLE
2 = SUBJECT IS NOT AVAILABLE (SET APPOINTMENT)
3 = WRONG NUMBER
4 = LANGUAGE BARRIER
5 = REFUSED
Q2_Intro - Informed Consent
You may remember completing a questionnaire for the Millennium Cohort Study on…
[FILL DATE/TIME & MODE OF LAST COMPLETED SURVEY]
I’d like to ask you a few of questions about why you have not yet completed a follow-up survey for
this study. This should take approximately 5 minutes. The information you provide will be shared
with the Millennium Cohort Study Team which is part of the Department of Defense. All of your
answers will be kept completely confidential. Your participation is voluntary, and you can stop at any
time.
1=Continue
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
A-1
Final Report
December 23, 2005
Non-Response Questions - Beginning of Blaise Questionnaire
Q1. What are the two main reasons you did not participate in this round of the study?
1 = I do not recall getting an invitation this round
2 = I could not access the website or I have no computer access
3 = I was deployed
4 = Survey was too long
5 = I’m out of the military now
6 = My health kept me from being able to do it
7 = I just wasn’t interested
8 = Other {RANGE 100 CHARACTERS}
9 = NO SECOND OPTION INDICATED
-1 = DK
-2 = RF
Q2. Are you aware that we are currently offering gifts to study members who continue to participate
in the study?
1 = Yes
2 = No
-1 = DK
-2 = RF
Q3. We currently offer our participants a 60-minute phone card, a hat, or a T-shirt for completing the
survey - all with the Millennium Cohort logo. Of the following items, which two do you think would
better encourage participation?
1 = Gift cards
2 = Coffee mugs
3 = Paid subscription to a magazine (If yes, continue to Q4)
4 = Clothing with service-specific logos
5 = Millennium Cohort coin
6 = Cash (If yes, continue to Q5)
7 = Something else {RANGE 100 CHARACTERS}
8 = NOTHING (GIFT IS NOT NECESSARY)
9 = DOES NOT MATTER (WOULD NOT DO SURVEY)
-1 = DK
-2 = RF
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
A-2
Final Report
December 23, 2005
{If Q3= 1, 2, 4, 5, 7, 8, 9, -1, -2, skip to Q6.}
Q4. What particular magazine do you think would encourage participation?
{RANGE 100 CHARACTERS}
{If Q3=3, skip to Q4}
Q5. What amount of cash do you think would encourage participation?
{RANGE $00.00 to $500.00}
{If Q3=6, skip to Q5}
Q6. In general, would you say your health is…
1 = Excellent
2 = Very Good
3 = Good
4 = Fair, or
5 = Poor
-1 = DK
-2 = RF
Participation Prompting
The current round of the survey is on-going and I’d like to encourage you to continue participating in
this important study.
Q7. Have you received any notifications in the mail about the study?
1 = Yes
2 = No
-1 = DK
-2 = RF
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
A-3
Final Report
December 23, 2005
Q8. Have you received any e-mails about the study?
1 = Yes
2 = No
3 = I don’t have an e-mail account
-1 = DK
-2 = RF
Q9. Do you have Internet access to complete the survey?
1 = Yes
2 = No
-1 = DK
-2 = RF
{If Q9= 1, skip to Q10. If Q9= 2, -1, -2, skip to Q12}
Q10. May I e-mail you a link to access the survey online?
1 = Yes
2 = No
-1 = DK
-2 = RF
{If Q10= 1, skip to Q11. If Q10= 2, -1, -2, skip to Q12}
Q11. What is your current e-mail address?
{RANGE 100 CHARACTERS}
Q12. Would you like us to mail you a paper version of the survey?
1 = Yes
2 = No
-1 = DK
-2 = RF
{If Q12= 1, skip to Q13. If Q12 = 2, -1, -2, skip to Q14}
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
A-4
Final Report
December 23, 2005
Q13. We have the following contact information for you.
•
NAME:
•
STREET ADDRESS:
•
CITY:
•
STATE:
•
ZIP:
Is this information still accurate?
1 = All information correct
2 = Need to change any piece of information
-1 = DK
-2 = RF
Q13a. You should receive the hardcopy questionnaire in approximately 1-2 weeks.
{If Q13 = 1, skip to Q13a. If Q13=2, skip to Q13 and correct address information. If q13= -1 or -2,
skip to Q14}
Completing the Full Survey Script
Q14. Thank you for answering those questions. Your participation is vital to helping the Department
of Defense evaluate the long term health of people who serve in the military. Do you have any
additional thoughts, questions, or comments that you would like to share concerning this study?
{RANGE 500 CHARACTERS}
Closing Script
Q15. Thank you for your time. Have a nice day/evening.
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
A-5
Final Report
December 23, 2005
Appendix B – Affidavit of
Non-Disclosure
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
Final Report
December 23, 2005
AFFIDAVIT OF NONDISCLOSURE
Millennium Cohort Study: Pilot Study of Non-Responders
(RTI Under Contract No. 05-D-2500 TO 002)
(Name)
(Job Title)
(Date of Assignment to Millennium Cohort Study)
Research Triangle Institute
(Organization, State or Local Agency or Instrumentality)
P.O. Box 12194, Research Triangle Park, NC 27709
(Address)
, do solemnly swear (or affirm) that when given access to any Millennium Cohort
I,
Study project-related data bases or files containing individually identifiable information, I will not:
(i)
use or reveal any individually identifiable information furnished, acquired, retrieved or assembled by me or
others, for any purpose other than statistical purposes in the Millennium Cohort Study or contract;
(ii)
make any disclosure or publication whereby a sample unit or survey respondent could be identified or the
data furnished by or related to any particular person under this section can be identified; or
(iii) permit anyone other than the individuals authorized by the Millennium Cohort Team to examine the
individual reports.
___________________________________
(Signature)
(The penalty for unlawful disclosure is a fine of not more than $250,000 (under 18 U.S.C. 3571) or imprisonment
for not more than 5 years (under 18 U.S.C. 3559), or both. The word "swear" should be stricken out wherever it
appears when a person elects to affirm the affidavit rather than to swear to it.)
State of North Carolina
County of ________________
I, ___________________________________, a Notary Public of _____________________ County, North Carolina certify
that _________________________________ personally appeared before me this day, and being duly sworn, stated that in
his/her presence _______________________________ signed the foregoing instrument.
Witness my hand and official seal, this the ___________ day of ________________, 2005.
___________________________________________
Notary Public
My Commission expires: ________________________
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
B-1
Final Report
December 23, 2005
Appendix C – Confidentiality
Agreement
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
Final Report
December 23, 2005
Millennium Cohort Study:
Pilot Study of Non-Responders Confidentiality Agreement
I, ___________________________________________ (print employee’s name), an employee of
Greene Resources, Inc., an independent contractor utilized by RTI, agree to work on all RTI
projects in accordance with the guidelines and restrictions specified below. I understand that
compliance with the terms of this agreement is a condition of my assignment with RTI and that these
terms are supplementary to those listed in my contract of employment with Greene Resources, Inc.
a. I agree to treat as confidential all case-specific information obtained in any RTI project and
related matters. I further agree that this covenant of confidentiality shall survive the
termination of this agreement.
b. I further understand that failure to follow the guidelines below may result in a potential
violation of the provisions of the Privacy Act of 1974 (violation of the Privacy Act is a
misdemeanor and may subject the violator to a fine of up to $5,000), and potential Institute
disciplinary action, including termination. To fulfill confidentiality obligations, I will:
1. Discuss confidential project information only with authorized employees of RTI.
2. Store confidential project information as specified by project protocols.
3. Safeguard combinations, keys, and rooms that secure confidential project
information.
4. Safeguard confidential project information when in actual use.
5. Immediately report any alleged violations of the security procedures to my immediate
supervisor.
6. Not photocopy or record by any other means any confidential project information
unless authorized by project leaders or my supervisor.
7. Not in any way compromise the confidentiality of project participants.
8. Not allow access to any confidential project information to any unauthorized person.
9. Report any lost or misplaced confidential project information to my supervisor
immediately.
_______________________________________________
Employee’s Signature
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
C-1
______/______/______
Date
Final Report
December 23, 2005
Appendix D – Debriefing Form
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
Final Report
December 23, 2005
Millennium Cohort Study: RTI-CCS Debriefing
Thursday, November 10th
The Millennium Cohort data collection period is quickly coming to an end. Thanks to your
continued efforts and dedication to the project, the project has been successful.
We need —and really want—your comments about how things are working or not working in
order to make improvements.
In advance of the debriefing, we ask that you review these questions and take notes about your
own experiences on the Millennium Cohort Study. These notes will help focus the conversation
during the debriefing. Please use the training materials, available from your supervisor, to jog
your memory and to help you be as specific as possible when referring to a particular question or
item in the instrument.
Please take no more than 15 minutes to write down your thoughts related to the questions listed
below and make any suggestions you have for how we can improve the study for the next wave.
Also, write down any other study related topics you’d like to discuss during the debriefing
session. We would like to collect these notes following the debriefing. Thank you in advance
and we look forward to seeing you!
Training:
1)
Do you think you were adequately prepared after the training session?
2)
What do you think we could improve for the next training?
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
D-1
Final Report
December 23, 2005
Tracing:
This section is for TOPS staff only.
1)
What tracing sources were most/least useful for intensive tracing?
2)
Did the tracers notice what preload information was most/least often correct?
3)
What preload information was most/least useful in intensive tracing?
4)
What was the average time needed to trace a case?
5)
What additional preload information would be helpful with intensive tracing?
Contacting/Locating Issues:
1)
Did we seem to have good locating information to start with?
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
D-2
Final Report
December 23, 2005
2)
Were contacts (places of work, family members of respondents) generally willing to
provide locating information?
3)
What problems did you see with finding or speaking with these military or no longer
active military staff?
4)
What did the respondents say was the best way to contact them (phone, e-mail, fax, cell
phone, office, through secretary)?
5)
Did you find that many respondents refused to complete the survey unless a paper copy
of the instrument was available?
Questionnaire Administration:
1)
How easy or difficult was it for you to use the Blaise instrument?
2)
Thinking about the flow of the interview – what worked well? What didn’t work well?
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
D-3
Final Report
December 23, 2005
3)
Thinking about the wording of questions – what questions were problematic and why?
What questions do you think were misunderstood by respondents and why?
4)
Were there sections or questions that the respondents felt were repetitious or
inappropriate?
5)
Did you use the training materials during the interview? Was the information useful?
Do any questions or terms need more explanation or more details? If so, which ones?
6)
Which items or questions in particular did you or respondents feel didn’t work well?
7)
At what point in the interview did the respondents typically complain?
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
D-4
Final Report
December 23, 2005
Additional Comments:
Please record any additional comments or questions you would like to bring up for the project
staff either during the debriefing or for the record. Thank you for your continued assistance!
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
D-5
Final Report
December 23, 2005
Appendix E – Frequency List
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
Final Report
December 23, 2005
Q1: 1st main reason not participate
Cumulative
Cumulative
Q1_1
Frequency
Percent
Frequency
Percent
-----------------------------------------------------------------------------------------------Doesn't recall getting invitation
158
17.40
158
17.40
Could not access website/no computer access
44
4.85
202
22.25
Was deployed
65
7.16
267
29.41
Survey was too long
51
5.62
318
35.02
Out of the military now
258
28.41
576
63.44
Health kept me from being able to do it
5
0.55
581
63.99
Just wasn't interested
86
9.47
667
73.46
Other
217
23.90
884
97.36
No 2nd opinion indicated
21
2.31
905
99.67
Refused
1
0.11
906
99.78
Don't know
2
0.22
908
100.00
Q1: 2nd main reason not participate
Cumulative
Cumulative
Q1_2
Frequency
Percent
Frequency
Percent
-----------------------------------------------------------------------------------------------.
147
.
.
.
Doesn't recall getting invitation
53
6.96
53
6.96
Could not access website/no computer access
30
3.94
83
10.91
Was deployed
12
1.58
95
12.48
Survey was too long
38
4.99
133
17.48
Out of the military now
65
8.54
198
26.02
Health kept me from being able to do it
6
0.79
204
26.81
Just wasn't interested
66
8.67
270
35.48
Other
160
21.02
430
56.50
No 2nd opinion indicated
331
43.50
761
100.00
Frequency Missing = 147
Q1 specify: First Other reason not participate
Q1_oth
DON'T WANT TO TELL
Did not have access to paperwork(out town)
Didnt receive feedback from first time, so
felt that it didnt need to be done again.
Does not remember receiving the study
Don't know
Going through divorce and is very busy.
Hadn't had time,and just got home
SM completed hard copy a couple of
months ago
SM did not receive t-shirt from last survey
SM moved and the information got lost
SM moved from his last location
SM said he did not remember receiving
study
SM said his wife discarded it
SM said she did mail in a hard copy back in
May but she wanted to complete this over
the phone anywa
SM said she forgot
SM said she lost hard copy
SM simply wasn't interested
SM thinks he completed it in June or July &
returned it.
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
Cumulative Cumulative
Frequency
Percent
1
0.46
2
0.92
Frequency
1
1
Percent
0.46
0.46
1
0.46
3
1.38
1
1
1
1
0.46
0.46
0.46
0.46
4
5
6
7
1.84
2.3
2.76
3.23
1
0.46
8
3.69
1
1
1
0.46
0.46
0.46
9
10
11
4.15
4.61
5.07
1
0.46
12
5.53
1
0.46
13
5.99
1
0.46
14
6.45
1
1
1
0.46
0.46
0.46
15
16
17
6.91
7.37
7.83
1
0.46
18
8.29
E-1
Final Report
December 23, 2005
Q1 specify: First Other reason not participate
Q1_oth
SM was in school
Son died and he is very busy now
TDY most of the time
Thought the questions were too personal
Time was the only reason
Too busy
Too busy.
Very Busy
Very busy
a little upset because she got a registered
mail and had to wait a long time at the post
office
alot of it didnt pertain 2 her, was never on
the battle field and knowingit was
voluntarry no rreaso
already completed web interview
already turn interview in
army recruiter
b/c he was getting out of the army
b/c not n the state
b/c she really go to the doctor that often
bad time in his life
bec i dont think it is doing any good
bec she asked several people if they heard
of the milcohort study -- military people -and they nev
because thought that it was voluntary
been oversees fight the war
busy
busy and working
busy at work
busy with family and he works a lot
busy, same thing he answered last time
career counseling and being real busy
change of address
co-workers or supervisor hadn't heard
about it so she thought it was a host or
identity theif
completed two surveys previously, tired of
surveys
could not complete survey because of code
being wrong
could not participate
death in the family
denied opportunity to take college classes
did not get around to it
did not get around to it or forgot
did not get it when he relocated
did not have the time
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
Cumulative Cumulative
Frequency
Percent
19
8.76
20
9.22
21
9.68
22
10.14
23
10.6
25
11.52
26
11.98
27
12.44
28
12.9
Frequency
1
1
1
1
1
2
1
1
1
Percent
0.46
0.46
0.46
0.46
0.46
0.92
0.46
0.46
0.46
1
0.46
29
13.36
1
0.46
30
13.82
1
1
1
1
1
1
1
1
0.46
0.46
0.46
0.46
0.46
0.46
0.46
0.46
31
32
33
34
35
36
37
38
14.29
14.75
15.21
15.67
16.13
16.59
17.05
17.51
1
0.46
39
17.97
1
1
1
1
1
1
1
1
1
0.46
0.46
0.46
0.46
0.46
0.46
0.46
0.46
0.46
40
41
42
43
44
45
46
47
48
18.43
18.89
19.35
19.82
20.28
20.74
21.2
21.66
22.12
1
0.46
49
22.58
1
0.46
50
23.04
1
0.46
51
23.5
1
1
1
1
1
1
1
0.46
0.46
0.46
0.46
0.46
0.46
0.46
52
53
54
55
56
57
58
23.96
24.42
24.88
25.35
25.81
26.27
26.73
E-2
Final Report
December 23, 2005
Q1 specify: First Other reason not participate
Q1_oth
did not know what it was about and thought
that it will be held against him later on
did survey on website & completed it and
thought it had been sent
didn't get around to it
didn't have time
didn't have time to sit down and do it
didn't know if it was important
didn't trust it and not in the military
didnt have enough time
didnt have time
didnt have time 2 do it...was n iraq and was
retired
didnt know had 2 keep doing it every couple
of years
didnt know if recieved
didnt know still could ...had retired in Jan.
didnt know what the study was
didnt know, b/c he moves alot didnt get
anything
didnt like the quest. (the same quest. over
annd over)
didnt recall seeing one
didnt recieve anything
didnt recieve it
doesnt have time
doesnt remember doing a study
doesnt remember doing survey
doesnt remember getting it
doesnt remember it
doesnt remember recieving anything
dont have time to complete the survey,
dont know just didnt think about it
dont remember
dont remember the study
e-mail it in already
filled out paper copy but is missing in the
home somewhere
first one she filed out they never
acknowledge it
getting out of mililtary
got a divorce and stuff got lost
got n the mail long ago and got lost
got too busy and forgot to do it
had other things 2 do
has been busy
has been busy and no time to do it
has been very busy preparing for a mission
has been very busy, has own business
has twin babies too busy
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
Frequency
E-3
Percent
Cumulative Cumulative
Frequency
Percent
1
0.46
59
27.19
1
0.46
60
27.65
1
1
1
1
1
1
1
0.46
0.46
0.46
0.46
0.46
0.46
0.46
61
62
63
64
65
66
67
28.11
28.57
29.03
29.49
29.95
30.41
30.88
1
0.46
68
31.34
1
0.46
69
31.8
1
1
1
0.46
0.46
0.46
70
71
72
32.26
32.72
33.18
1
0.46
73
33.64
1
0.46
74
34.1
1
2
1
1
1
1
1
1
1
1
1
1
1
1
0.46
0.92
0.46
0.46
0.46
0.46
0.46
0.46
0.46
0.46
0.46
0.46
0.46
0.46
75
77
78
79
80
81
82
83
84
85
86
87
88
89
34.56
35.48
35.94
36.41
36.87
37.33
37.79
38.25
38.71
39.17
39.63
40.09
40.55
41.01
1
0.46
90
41.47
1
0.46
91
41.94
1
1
1
1
1
1
1
1
1
1
0.46
0.46
0.46
0.46
0.46
0.46
0.46
0.46
0.46
0.46
92
93
94
95
96
97
98
99
100
101
42.4
42.86
43.32
43.78
44.24
44.7
45.16
45.62
46.08
46.54
Final Report
December 23, 2005
Q1 specify: First Other reason not participate
Q1_oth
hasnt got it
hasnt gotten around to filling out the survey
haven't had time
he did complete the survey and sent it in
he did it on line -- rc
he did send in the survey
he did the survey -- august 16, 2005
he did the survey on line -- may or june 2005
he did the survey on the internet and did not
receive his incentive --- will do again -he does not take it seriously
he feels like the survey does not fit his job
description
he filled it out and got a letter in the mail
and he filled it out and turned it in.
he forgot about it
he has been recalled 3x since first study;
lapse in receipt of mail and other things
he has completed them and he keeps
getting them. He gets them every three
months
he sent it in june
he was busy and he misplaced it.
hethought he did it and did not turn it in
his mother passed away
hurricane
hurricane katrina
in the process of moving
involved in a contstruction project, not high
on priority list
job with lots of traveling and very busy,
just didnt get it , moved alot n the past
years
just didnt get the survey
just didnt have alot of time
just didnt have the time, didnt get anything,
just moved
just didnt have time
just didnt see a reason 2 do survey
just don't have time and thought it was a
one time thing
just forgot 2 get to it
just has been busy
just never did it
just procrastinated, had planned on it
just too busy
lost paper work
lost the invitation
lost the paper work and forgot
mailed packet and it was lost
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
Cumulative Cumulative
Frequency
Percent
102
47
103
47.47
104
47.93
105
48.39
106
48.85
107
49.31
108
49.77
109
50.23
Frequency
1
1
1
1
1
1
1
1
Percent
0.46
0.46
0.46
0.46
0.46
0.46
0.46
0.46
1
0.46
110
50.69
1
0.46
111
51.15
1
0.46
112
51.61
1
0.46
113
52.07
1
0.46
114
52.53
1
0.46
115
53
1
0.46
116
53.46
1
1
1
1
1
1
1
0.46
0.46
0.46
0.46
0.46
0.46
0.46
117
118
119
120
121
122
123
53.92
54.38
54.84
55.3
55.76
56.22
56.68
1
0.46
124
57.14
1
0.46
125
57.6
1
0.46
126
58.06
1
1
0.46
0.46
127
128
58.53
58.99
1
0.46
129
59.45
2
1
0.92
0.46
131
132
60.37
60.83
1
0.46
133
61.29
1
1
1
1
1
1
1
1
1
0.46
0.46
0.46
0.46
0.46
0.46
0.46
0.46
0.46
134
135
136
137
138
139
140
141
142
61.75
62.21
62.67
63.13
63.59
64.06
64.52
64.98
65.44
E-4
Final Report
December 23, 2005
Q1 specify: First Other reason not participate
Q1_oth
maybe just forgot about it
mislaced survey
misplace survey
missplaced survey
mom has been in the hospital for the last 2
months
most studies that the military do does they
do not pay attention to do it.
moved
moved change of address
moving around alot
my job keeps me too busy and i'm too busy
at home doing the same things.
never got any info. on it
never receive a response/never receive gift
never recieved survey
no
no other reason given
no time
not enough time
not interested
out of military since 01 & wanted current
physical bf completing form
personal problems and just did not want to
do it
probably just layed it down and didnt look
at it agagin
questions were way 2 personal, asking
unfo. that he just doesnt giveout
really doesnt have a resason
really dont recall doing study
received e-mail in error
received the information but misunderstood
focus of the study did not realize was a long
term study
retired
sent survey in
sent survey in within the last two weeks
she just mailed and in school
shift change
sm said completed via e-mail in August
2005.
sm was tdy
started working on questionnaire and was
unable to complete it
swamped with transition out of navy
thinks he completed it
this lady already did the study on line two
times in the last year
thought already done it
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
Cumulative Cumulative
Frequency
Percent
143
65.9
144
66.36
145
66.82
146
67.28
Frequency
1
1
1
1
Percent
0.46
0.46
0.46
0.46
1
0.46
147
67.74
1
0.46
148
68.2
3
1
1
1.38
0.46
0.46
151
152
153
69.59
70.05
70.51
1
0.46
154
70.97
1
1
1
1
1
2
1
1
0.46
0.46
0.46
0.46
0.46
0.92
0.46
0.46
155
156
157
158
159
161
162
163
71.43
71.89
72.35
72.81
73.27
74.19
74.65
75.12
1
0.46
164
75.58
1
0.46
165
76.04
1
0.46
166
76.5
1
0.46
167
76.96
1
1
1
0.46
0.46
0.46
168
169
170
77.42
77.88
78.34
1
0.46
171
78.8
1
1
1
1
1
0.46
0.46
0.46
0.46
0.46
172
173
174
175
176
79.26
79.72
80.18
80.65
81.11
1
0.46
177
81.57
1
0.46
178
82.03
1
0.46
179
82.49
1
1
0.46
0.46
180
181
82.95
83.41
1
0.46
182
83.87
1
0.46
183
84.33
E-5
Final Report
December 23, 2005
Q1 specify: First Other reason not participate
Q1_oth
thought had already done survey
thought he had some important mail and
went to the post office for certified mail and
had to go twic
thought it was junk mail
thought she filled oput survey
thought the survey was repetitive
thought there was a deadline and he was
busy
time
time constraint
to busy and retired moved to another home
and in school
to many e-mails from milcohort
too busy
too busy.
too little time
took on new career,
traveling alot
tryed to completed a couple of times and
system kicked database back.
very busy and also not knowing there was
another study such as this
very busy and did not get around to it.
wanted to know frequency of survey &
wanted to talk w/person, not website
was out of military then and didnt think it
applied to him
was out of the country working not in the
military
was pending discharged , didnt know could
do it
was unware of study follow up
wasn't aware about it being a continue
wasnt a particular reason, just didnt
way to busy and about to get deployed and
the last thing on his list
we are surveyed so many times that we do
not know what is good
when he goes to reserve site, too many
surveys & tired of doing them
work 6 days a wk 12hrs a day
worked related issues
you did mail in the questionnaire about 1
week after receiving
you never got around to doing it
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
Cumulative Cumulative
Frequency
Percent
184
84.79
Frequency
1
Percent
0.46
1
0.46
185
85.25
1
1
1
0.46
0.46
0.46
186
187
188
85.71
86.18
86.64
1
0.46
189
87.1
2
1
0.92
0.46
191
192
88.02
88.48
1
0.46
193
88.94
1
2
1
1
1
1
0.46
0.92
0.46
0.46
0.46
0.46
194
196
197
198
199
200
89.4
90.32
90.78
91.24
91.71
92.17
1
0.46
201
92.63
1
0.46
202
93.09
1
0.46
203
93.55
1
0.46
204
94.01
1
0.46
205
94.47
1
0.46
206
94.93
1
0.46
207
95.39
1
1
1
0.46
0.46
0.46
208
209
210
95.85
96.31
96.77
1
0.46
211
97.24
1
0.46
212
97.7
1
0.46
213
98.16
1
1
0.46
0.46
214
215
98.62
99.08
1
0.46
216
99.54
1
0.46
217
100
E-6
Final Report
December 23, 2005
Q1 specify: Second Other reason not participate
Q1_oth
Had school finals
Moved recently
No time
SM did not know that he could still
participate
SM said he feels that the survey won't do any
good
SM was down on the military
SM was very busy when he use to respond to
the questionnaires
SM's job keeps him from doing other things
Some of questions were too personal, not
applicable and not comfortable.
TDY alot
Thought that survey was done
Tired of doing the survey and hasnt seen any
changes, so he thinks it is about making
someone feel s
Very busy.
Was very busy.
active military and busy with school
almost positive that he mailed survey back in
April
always busy
b/c we havent sent him anytihng
back n the militray and just doesnt want to do
it anymore
bedrest while pregnant
busy
busy and don't have time to complete
busy and forgot
busy schedule
busy then
busyt
couldn't find the paper
did it in 2002
did not have time
did not know how long survey was going on
did not know you could do on computer
did not receive information
did not remember receiving paper work
did not think this was mandatory
didn't have the time
didn't have time
didn't have time, laziness
didn't know about it
didn't know he could do it online
didn't know it was from dod
didn't know there was a follow-up survey
didn't want to take time.
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
Frequency
1
1
1
E-7
Cumulative Cumulative
Percent
Frequency
Percent
0.63
1
0.63
0.63
2
1.25
0.63
3
1.88
1
0.63
4
2.5
1
0.63
5
3.13
1
0.63
6
3.75
1
0.63
7
4.38
1
0.63
8
5
1
0.63
9
5.63
1
1
0.63
0.63
10
11
6.25
6.88
1
0.63
12
7.5
1
1
1
0.63
0.63
0.63
13
14
15
8.13
8.75
9.38
1
0.63
16
10
1
1
0.63
0.63
17
18
10.63
11.25
1
0.63
19
11.88
1
8
1
1
1
1
1
1
1
2
1
1
1
1
1
1
2
1
1
1
1
1
1
0.63
5
0.63
0.63
0.63
0.63
0.63
0.63
0.63
1.25
0.63
0.63
0.63
0.63
0.63
0.63
1.25
0.63
0.63
0.63
0.63
0.63
0.63
20
28
29
30
31
32
33
34
35
37
38
39
40
41
42
43
45
46
47
48
49
50
51
12.5
17.5
18.13
18.75
19.38
20
20.63
21.25
21.88
23.13
23.75
24.38
25
25.63
26.25
26.88
28.13
28.75
29.38
30
30.63
31.25
31.88
Final Report
December 23, 2005
Q1 specify: Second Other reason not participate
Q1_oth
Frequency
didnt c benefit n completing the study, no
1
matter what...nothing changes(medical field
has declined)
1
didnt think the questions pertained 2 him
disgruntled because never received a hat as
1
promised during 1st phase of research
1
does not remember first round
1
doesnt have time
1
don't feel like doing it anymore
1
don't have enough time
1
don't remember getting it
1
don't think the surveys do any good, too long
1
education was taken away, fought in 2 wars
every time he will try to do the survey he got
1
error messages
first time I did it, thought it was worthless,
1
getting ready to deploy and too many other
important
1
forgot about it
1
forgot busy
forgot to mail the survey left it on pation and
1
it got wet
found that surveys don't accomplish
1
anything from experience his supervisor nevr
cared and didn't do
1
going thru divorce -- busy
1
had no time
1
hardly have time
1
has been very busy
1
has moved 3 times
1
has no time , works two jobs
1
hasnt recieved any more info on it
1
hate surveys
1
haven't had time
1
havent recieved one
1
he did it already
1
he doesn't know
2
he forgot about it
1
he lost track of it in hectic house
1
he might do it when he got time
1
he moved
he tried to access the website many times
1
and it won't let me access it
1
husband ill has been to busy
i disagree with everything and invasion of
1
innocent people
1
in navy school
1
in the process of moving
in various locations & prob has missed some
1
mail
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
E-8
Percent
Cumulative Cumulative
Frequency
Percent
0.63
52
32.5
0.63
53
33.13
0.63
54
33.75
0.63
0.63
0.63
0.63
0.63
0.63
0.63
55
56
57
58
59
60
61
34.38
35
35.63
36.25
36.88
37.5
38.13
0.63
62
38.75
0.63
63
39.38
0.63
0.63
64
65
40
40.63
0.63
66
41.25
0.63
67
41.88
0.63
0.63
0.63
0.63
0.63
0.63
0.63
0.63
0.63
0.63
0.63
0.63
1.25
0.63
0.63
0.63
68
69
70
71
72
73
74
75
76
77
78
79
81
82
83
84
42.5
43.13
43.75
44.38
45
45.63
46.25
46.88
47.5
48.13
48.75
49.38
50.63
51.25
51.88
52.5
0.63
85
53.13
0.63
86
53.75
0.63
87
54.38
0.63
0.63
88
89
55
55.63
0.63
90
56.25
Final Report
December 23, 2005
Q1 specify: Second Other reason not participate
Q1_oth
Frequency
1
is not seeing a result
just couldn't remember what the study was
1
about or who was conducting it
1
just did not do it
2
just didn't do it
1
just didnt get around to it
1
just didnt get to it
1
just didnt have time
just have been busy and doesnt have internet
1
at sea
just kind of put it down, havent gotten back
1
to it
1
lack of time
1
lazy
1
lost interest
1
lost password and website
1
mailed survey in
1
may have deleted e mail
1
moved so much it might not came to her
1
moving around a lot in the last few years
1
never been in the military
1
never get anything n mail
1
never received gift
1
no longer at same address
2
no other reason
2
no time
1
no time is in National Guard full time
1
not enough time
1
not interested
1
out of town alot with new job
1
personal problems
1
probably sent the paperwork in
1
procastinating
1
process of pcs and transfering
1
procrastination
1
questions don't make sense.
1
really busy
1
really busy, has a second job, doing trucking
retired in April and very busy with moving
1
and other things
selling one house and buying another and
1
busy all the time
she has been out since 2001 and feels it
1
would not be related
1
so busy
1
someone told hi
1
subj sd she doesn't have time
1
survey is not helpful
1
thought it was too late when she got back
1
time constraints
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
E-9
Cumulative Cumulative
Percent
Frequency
Percent
0.63
91
56.88
0.63
92
57.5
0.63
1.25
0.63
0.63
0.63
93
95
96
97
98
58.13
59.38
60
60.63
61.25
0.63
99
61.88
0.63
100
62.5
0.63
0.63
0.63
0.63
0.63
0.63
0.63
0.63
0.63
0.63
0.63
0.63
1.25
1.25
0.63
0.63
0.63
0.63
0.63
0.63
0.63
0.63
0.63
0.63
0.63
0.63
101
102
103
104
105
106
107
108
109
110
111
112
114
116
117
118
119
120
121
122
123
124
125
126
127
128
63.13
63.75
64.38
65
65.63
66.25
66.88
67.5
68.13
68.75
69.38
70
71.25
72.5
73.13
73.75
74.38
75
75.63
76.25
76.88
77.5
78.13
78.75
79.38
80
0.63
129
80.63
0.63
130
81.25
0.63
131
81.88
0.63
0.63
0.63
0.63
0.63
0.63
132
133
134
135
136
137
82.5
83.13
83.75
84.38
85
85.63
Final Report
December 23, 2005
Q1 specify: Second Other reason not participate
Q1_oth
too busy
too busy @ work
too busy moving and doing day to day stuff
too many surveys and got tired and
disinterested
too personal
took a new job & very busy with commuting
& other
tooo busy
very busy
very busy and did not have time
very busy in deployment
very busy with on job
was moved and havent got another one
was moving and had computer down
was very busy for a long time
when completing the lst survey, did not
realize there going to be subsequent surveys
you send too many surveys in the mail
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
Frequency
6
1
1
E-10
Cumulative Cumulative
Percent
Frequency
Percent
3.75
143
89.38
0.63
144
90
0.63
145
90.63
1
0.63
146
91.25
1
0.63
147
91.88
1
0.63
148
92.5
1
3
1
1
1
1
1
1
0.63
1.88
0.63
0.63
0.63
0.63
0.63
0.63
149
152
153
154
155
156
157
158
93.13
95
95.63
96.25
96.88
97.5
98.13
98.75
1
0.63
159
99.38
1
0.63
160
100
Final Report
December 23, 2005
Q2: Aware of incentive offers?
Cumulative
Cumulative
Q2
Frequency
Percent
Frequency
Percent
--------------------------------------------------------------Yes
322
35.46
322
35.46
No
579
63.77
901
99.23
Refused
1
0.11
902
99.34
Don't know
6
0.66
908
100.00
Q3: 1st item encourage participation
Cumulative
Cumulative
Q3_1
Frequency
Percent
Frequency
Percent
----------------------------------------------------------------------------------------Gift cards
263
28.96
263
28.96
Coffee mugs
57
6.28
320
35.24
Paid subscription to magazine
84
9.25
404
44.49
Clothing with service-specific logo
102
11.23
506
55.73
Millennium Cohort coin
41
4.52
547
60.24
Cash
239
26.32
786
86.56
Something else, specify
22
2.42
808
88.99
Nothing - gifts not necessary
38
4.19
846
93.17
Does not matter, would not do survey
45
4.96
891
98.13
Refused
1
0.11
892
98.24
Don't know
16
1.76
908
100.00
The FREQ Procedure
Q3: 2nd item encourage participation
Cumulative
Cumulative
Q3_2
Frequency
Percent
Frequency
Percent
----------------------------------------------------------------------------------------.
164
.
.
.
Gift cards
139
18.68
139
18.68
Coffee mugs
67
9.01
206
27.69
Paid subscription to magazine
82
11.02
288
38.71
Clothing with service-specific logo
118
15.86
406
54.57
Millennium Cohort coin
38
5.11
444
59.68
Cash
165
22.18
609
81.85
Something else, specify
36
4.84
645
86.69
Nothing - gifts not necessary
57
7.66
702
94.35
Does not matter, would not do survey
42
5.65
744
100.00
Frequency Missing = 164
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
E-11
Final Report
December 23, 2005
Q3 specify: First Other item encourage participation
Cumulative Cumulative
Q3_oth
Frequency
Percent
Frequency
Percent
1
4.55
1
4.55
Don't know
NONE OF THE ABOVE HE THINKS PEOPLE
1
4.55
2
9.09
WOULDN'T DO IT
SM said to let the subjects choose what they
1
4.55
3
13.64
want
1
4.55
4
18.18
all of the choices are pretty good
1
4.55
5
22.73
calling card
1
4.55
6
27.27
fill out survey and them alone
1
4.55
7
31.82
gift card 2 a restaurant
1
4.55
8
36.36
give people a chose of those
1
4.55
9
40.91
make it shorter
1
4.55
10
45.45
nothing is appealing to him
1
4.55
11
50
other
2
9.09
13
59.09
phone card
3
13.64
16
72.73
phone cards
1
4.55
17
77.27
phonecard
1
4.55
18
81.82
shorten survey
1
4.55
19
86.36
th eamerican flag
the t shirt was ok -- the incentive was not
1
4.55
20
90.91
necessary but just to do it so it would help
peopleisok
1
4.55
21
95.45
ticket to sea world
would like to see result of the impact on the
1
4.55
22
100
constitution, improvements and would like to
see what
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
E-12
Final Report
December 23, 2005
Q3 specify: Second Other item encourage participation
Cumulative Cumulative
Q3_oth
Frequency
Percent
Frequency
Percent
1
2.78
1
2.78
5
1
2.78
2
5.56
Gas Card
1
2.78
3
8.33
Pen and pencil set
1
2.78
4
11.11
Phone Card
1
2.78
5
13.89
T-shirt
1
2.78
6
16.67
a cap
1
2.78
7
19.44
another shirt
1
2.78
8
22.22
cars
1
2.78
9
25
certificate
1
2.78
10
27.78
description on what participates are use for
1
2.78
11
30.56
football
1
2.78
12
33.33
gift card to walmart
1
2.78
13
36.11
gift certificate
1
2.78
14
38.89
gift certificate to restaurant
1
2.78
15
41.67
it would depend
1
2.78
16
44.44
more education
12
33.33
28
77.78
phone card
7
19.44
35
97.22
phone cards
1
2.78
36
100
the t-shirts
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
E-13
Final Report
December 23, 2005
Q4: Specify magazine to encourage participation
Q4
Army Times
Army Times Newspaper
Army Times, Newsweek, Business Week
Computer magazine
Cosmo
Cosmopolitan for women, GQ for men.
DOD related
Don't know
EBONY, JET, BLACK ENTRENPRENURS
Espn
Food & Wine
Golf Magazine
Health & Fitness
Health Magazine
Home/Family magazines
House Beautiful
IRON HORSE OR RIDER
Martha Stewart Living, Better Homes &
Gardens
Maxim
Maxim Magazine
McGraw Hill Magazine or McMillian Magazine
Men's Health
Men's health magazine
Military magazine
Military type magazine
National Geographics
Outdoor magazine
Parenting or fitness magazine
People
People Magazine
Popular Science
Smart Money
Sports Illstruated
Sports Illustrated
Sports Illustrated Magazine
Sports Illustrated or Men's health
Sports Illustrated, Men's Fitness
Sports magazine
Time Magazine
Time or Newsweek
a fitness magazine
a major one
air entertainment weekly
air force - afa? ncoa
air force times
air force times or army times
american woodworker
appropriate branch times magazine
automotive
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
Frequency
1
1
1
1
1
1
1
18
1
1
1
1
1
1
1
1
1
1
Percent
0.6
0.6
0.6
0.6
0.6
0.6
0.6
10.84
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
3
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
1.81
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
E-14
Cumulative Cumulative
Frequency
Percent
1
0.6
2
1.2
3
1.81
4
2.41
5
3.01
6
3.61
7
4.22
25
15.06
26
15.66
27
16.27
28
16.87
29
17.47
30
18.07
31
18.67
32
19.28
33
19.88
34
20.48
35
21.08
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
21.69
22.29
22.89
23.49
24.1
24.7
25.3
25.9
26.51
27.11
27.71
28.31
28.92
29.52
30.12
31.93
32.53
33.13
33.73
34.34
34.94
35.54
36.14
36.75
37.35
37.95
38.55
39.16
39.76
40.36
40.96
Final Report
December 23, 2005
Q4: Specify magazine to encourage participation
Q4
better homes and gardens
business weekly
car and driver
car and driver and road and track and
popular science
car and driver or road & track
consumer report
consumer reports
cosmo
decorating magazines, southern living
depends on individual
depends on person
discover
discovery
duck, or delta
economist
either service members mags navy times -service specific magazines
essence
everyday people or something
excellence
field and stream
field and stream, big buck, or the bass
magazines, sports illustrated
fitness
four wheeler
glamour, lucky
guns and ammo magazine
home improvement
hunting and fishing
marine corps gazette
mens fitness
mens health
military magazine
military magazine like Army Times
national review
navy times
news magazine
news week
newsweek or time or something like that
no choice
no idea
not interested in magazine
offer a choice
outdoor like hunting, fishing, gun magazine
pc or technical magazine
people
people magazine
people or life
per sm unknown
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
Cumulative Cumulative
Frequency
Percent
69
41.57
70
42.17
71
42.77
72
43.37
Frequency
1
1
1
1
Percent
0.6
0.6
0.6
0.6
1
1
1
1
1
1
1
1
1
1
1
1
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
73
74
75
76
77
78
79
80
81
82
83
84
43.98
44.58
45.18
45.78
46.39
46.99
47.59
48.19
48.8
49.4
50
50.6
1
1
1
2
1
0.6
0.6
0.6
1.2
0.6
85
86
87
89
90
51.2
51.81
52.41
53.61
54.22
2
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1.2
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
55.42
56.02
56.63
57.23
57.83
58.43
59.04
59.64
60.24
60.84
61.45
62.05
62.65
63.25
63.86
64.46
65.06
65.66
66.27
66.87
67.47
68.07
68.67
69.28
69.88
70.48
E-15
Final Report
December 23, 2005
Q4: Specify magazine to encourage participation
Q4
playboy
political
popular mechanics, popular science
popular science
popular science or popular mechanics
readers digest etc
remote control aircraft
retired military magazine
send publication list to choose
something 2 do with the military
sporting magazine
sports
sports illustrated
sports illustrated or reader's digest
sports illustrated, self
sports mag or cars for men, beauty for
women
sports magazine
sports magazine or military magazine
sports magazine, family circke
sports or business
sports, cars magazine
time
time magazine, sports illustrated
time or newsweek
times
tv guide---psychology today
unspecific
us news
what ever persons likes
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
Frequency
1
1
1
1
1
1
1
1
1
1
1
7
11
1
1
1
Percent
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
4.22
6.63
0.6
0.6
0.6
2
1
1
1
1
2
1
2
2
1
1
1
1
1.2
0.6
0.6
0.6
0.6
1.2
0.6
1.2
1.2
0.6
0.6
0.6
0.6
E-16
Cumulative Cumulative
Frequency
Percent
118
71.08
119
71.69
120
72.29
121
72.89
122
73.49
123
74.1
124
74.7
125
75.3
126
75.9
127
76.51
128
77.11
135
81.33
146
87.95
147
88.55
148
89.16
149
89.76
151
152
153
154
155
157
158
160
162
163
164
165
166
90.96
91.57
92.17
92.77
93.37
94.58
95.18
96.39
97.59
98.19
98.8
99.4
100
Final Report
December 23, 2005
Q5: Specify cash amount to encourage participation
Cumulative
Cumulative
Q5
Frequency
Percent
Frequency
Percent
------------------------------------------------------------.
504
.
.
.
Don'know
49
.
.
.
0
7
1.97
7
1.97
2
1
0.28
8
2.25
5
19
5.35
27
7.61
10
63
17.75
90
25.35
15
5
1.41
95
26.76
20
65
18.31
160
45.07
25
62
17.46
222
62.54
30
5
1.41
227
63.94
40
4
1.13
231
65.07
50
69
19.44
300
84.51
65
1
0.28
301
84.79
75
1
0.28
302
85.07
100
28
7.89
330
92.96
200
3
0.85
333
93.80
203
1
0.28
334
94.08
250
2
0.56
336
94.65
500
19
5.35
355
100.00
Frequency Missing = 553
Q6: Health rating
Cumulative
Cumulative
Q6
Frequency
Percent
Frequency
Percent
--------------------------------------------------------------Excellent
225
24.78
225
24.78
Very Good
309
34.03
534
58.81
Good
264
29.07
798
87.89
Fair
87
9.58
885
97.47
Poor
19
2.09
904
99.56
Refused
3
0.33
907
99.89
Don't know
1
0.11
908
100.00
Q7: Received any mail notifications?
Cumulative
Cumulative
Q7
Frequency
Percent
Frequency
Percent
--------------------------------------------------------------Yes
357
39.32
357
39.32
No
506
55.73
863
95.04
Refused
1
0.11
864
95.15
Don't know
44
4.85
908
100.00
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
E-17
Final Report
December 23, 2005
Q8: Received any e-mail notifications?
Cumulative
Cumulative
Q8
Frequency
Percent
Frequency
Percent
----------------------------------------------------------------------------Yes
134
14.76
134
14.76
No
678
74.67
812
89.43
Don't have e-mail account
35
3.85
847
93.28
Refused
1
0.11
848
93.39
Don't know
60
6.61
908
100.00
Q9: Have Internet access to complete survey?
Cumulative
Cumulative
Q9
Frequency
Percent
Frequency
Percent
--------------------------------------------------------------Yes
758
83.48
758
83.48
No
147
16.19
905
99.67
Refused
2
0.22
907
99.89
Don't know
1
0.11
908
100.00
Q10: Ok to e-mail online survey link?
Cumulative
Cumulative
Q10
Frequency
Percent
Frequency
Percent
--------------------------------------------------------------.
150
.
.
.
Yes
573
75.59
573
75.59
No
183
24.14
756
99.74
Refused
1
0.13
757
99.87
Don't know
1
0.13
758
100.00
Frequency Missing = 150
Q12: Request paper version
Cumulative
Cumulative
Q12
Frequency
Percent
Frequency
Percent
--------------------------------------------------------------Yes
308
33.92
308
33.92
No
599
65.97
907
99.89
Refused
1
0.11
908
100.00
Q13:
Contact information correct?
Cumulative
Cumulative
Q13
Frequency
Percent
Frequency
Percent
--------------------------------------------------------------------------------.
601
.
.
.
All information is correct
217
70.68
217
70.68
Need to change piece of info
90
29.32
307
100.00
Frequency Missing = 6
Millennium Cohort Study:
Telephone Study of 3,000 Non-Responders
E-18
Final Report
December 23, 2005
| File Type | application/pdf |
| File Title | Millennium Cohort Study: Telephone Study of 3,000 Non-Responders |
| Subject | MilCohort NonResponse Study Final Report |
| Author | RTI International for Naval Health Research Center |
| File Modified | 2005-12-29 |
| File Created | 2005-12-23 |