2011-12 SCHOOLS AND STAFFING SURVEY (SASS)

Status: Original and derived artifacts are available for this document.

Download: doc | pdf | html




2011-12 Schools and Staffing Survey (SASS)
PRELIMINARY FIELD ACTIVITIES





OMB Supporting Statement

Part B





OMB# 1850-0598 v.6











December 2010

National Center For Education Statistics (NCES)
B.	COLLECTION OF INFORMATION EMPLOYING STATISTICAL METHODS

    1. RESPONDENT UNIVERSE
        1.1. SCHOOLS
The respondent universe for the 2011-12 full-scale data collection consists of 95,750 public and 23,838 private schools in the 50 U.S. states and the District of Columbia (DC) that offer instruction in any of grades K-12.  To be eligible for inclusion in the sample, schools must provide classroom instruction to students, have one or more teachers to provide instruction, serve students in at least one of grades 1-12 or the ungraded equivalent, must be located in one or more buildings, and must be located in the U.S. and not in the outlying areas or U.S. territories.  No Bureau of Indian Affairs-funded schools will be included in the sampling frame for the 2011-12 SASS.

NCES’ 2008-09 Common Core of Data (CCD) and 2007-08 Private School Survey (PSS) will be used to construct the public and private school sampling frames, respectively, unless a more recent CCD file becomes available.  The respondent universe for charter schools will be identified as those public charter schools that meet the SASS definition of an eligible school found on the CCD.  Tables 4 and 5 below present the number of public schools on the 2008-09 CCD and the number of private schools on the 2009-10 PSS by region and school level.  The universe has been adjusted to remove K-terminal schools, which are not eligible for SASS. 

Table 4.	Respondent universe by school level and region for the proposed public school sample, based on the 2008-09 CCD
 
School level
Total
Region
Primary
Middle
High
Combined

Northeast
8,590
2,689
2,924
796
14,999
Midwest
13,497
4,404
6,205
1,565
25,671
South
17,368
6,207
6,381
2,582
32,538
West
12,526
3,492
4,871
1,653
22,542
Total
51,981
16,792
20,381
6,596
95,750
SOURCE: 2008-09 CCD.

Table 5.	Respondent universe by school level and region for the proposed private school sample, based on the 2007-08 PSS

School level
Total
Region
Elementary
Secondary
Combined

Northeast
3,381
797
1,438
5,616
Midwest
4,298
575
1,125
5,998
South
3,612
544
3,176
7,332
West
2,955
551
1,386
4,892
Total
14,246
2,467
7,125
23,838
SOURCE: 2007-08 PSS.

Details of the first-stage sample design of schools are provided in section 2.

        1.2. TEACHERS
Teachers will be randomly sampled within the second design stage from roster information provided by each participating sampled school.  Teachers within the sampled school are classified as ineligible for SASS if they are a short-term substitute teacher, student teacher, a teacher’s aide, or do not teach any of grades K-12 or comparable ungraded levels.  This information is obtained from the Teacher Questionnaire.   Details of the second-stage sample design of teachers are provided in section 2.

    2. STATISTICAL PROCEDURES FOR COLLECTING INFORMATION
        2.1. SAMPLING
The final 2011-12 SASS samples will include no more than:
    • 5,800 LEAs; 
    • 16,350 schools and school principals (12,000 traditional public, 750 charter, and 3,600 private schools); and
    • 74,900 teachers (58,700 traditional public, 3,800 charter, and 12,400 private school teachers).  

Sampling – Public Schools
The level of precision achieved by the 2007-08 SASS was evaluated to inform the sample design decisions for the 2011-12 SASS.  The precision analysis was based upon important analysis variables as well as generic proportions to address other important SASS characteristics.  These variables and values were evaluated: 
    • by state (public schools) and affiliation (private schools);
    • by school type (public charter, traditional public, overall public, private); 
    • by region within school type; 
    • by grade level (elementary, secondary, and combined for private schools; primary, middle, high, and combined for public schools);
    • by grade level and region within school type; and
    • by teachers’ years of experience.

The desired level of precision for SASS estimates was defined in terms of the estimated coefficient of variation (CV; also referred to as a relative standard error).  Three levels of precision were evaluated: 15, 20, and 30 percent CV.  

Table 6 presents a portion of the analysis for traditional public schools by state and public charter schools, highlighting the minimum respondent sample size needed for the 2011-12 SASS to achieve either a 15, 20, or 30 percent CV for two key estimates—the number of schools and the number of students.

Table 6. Minimum respondent sample size needed for 2011-12 SASS by percent CV for public schools by state and public charter schools


 
SASS 07-08
Frame
Count
SASS 07-08 Responding
Schools
Minimum Respondent Sample Size



Max %CV for Number of Schools Estimate
Max %CV for Number of Students Estimate

Max % CV
School type


15
20
30
15
20
30













Traditional Public by State










  Alabama
1,578
167
11
6
3
31
18
8

31
  Alaska
488
131
38
22
10
180
109
51

180
  Arizona
1,555
156
30
17
7
34
19
8

34
  Arkansas
963
154
11
6
3
18
10
4

18
  California
9,249
292
9
5
2
51
28
13

51
  Colorado
1,463
127
9
5
2
13
7
3

13
  Connecticut
1,059
115
6
3
2
25
14
6

25
  Delaware
210
90
149
97
49
79
48
22

149
  District of Columbia
174
75
18
10
5
34
20
9

34
  Florida
3,278
186
10
6
3
29
16
7

29
  Georgia
2,357
137
6
3
2
12
7
3

12
  Hawaii
257
77
117
72
34
44
25
12

117
  Idaho
672
154
33
18
8
48
27
12

48
  Illinois
4,112
147
4
2
1
19
11
5

19
  Indiana
1,908
140
15
9
4
15
9
4

15
  Iowa
1,313
130
3
2
1
29
16
7

29
  Kansas
1,334
137
61
35
16
33
19
8

61
  Kentucky
1,444
134
12
7
3
24
14
6

24
  Louisiana
1,469
127
19
11
5
34
19
9

34
  Maine
668
141
3
2
1
22
12
5

22
  Maryland
1,398
100
3
2
1
14
8
3

14
  Massachusetts
1,744
129
1
0
0
11
6
3

11
Table 6. Minimum respondent sample size needed for 2011-12 SASS by percent CV for public schools by state and public charter schools—Continued 
 
SASS 07-08
Frame
Count
SASS 07-08 Responding
Schools
Minimum Respondent Sample Size



Max %CV for Number of Schools Estimate
Max %CV for Number of Students Estimate

Max % CV
School type


15
20
30
15
20
30


  Michigan
3,614
136
11
6
3
34
19
8

34
  Minnesota
2,133
204
22
12
5
86
49
22

86
  Mississippi
1,044
153
1
1
0
12
7
3

12
  Missouri
2,002
186
12
7
3
19
10
5

19
  Montana
568
163
12
7
3
25
14
6

25
  Nebraska
1,041
137
31
18
8
28
16
7

31
  Nevada
528
124
21
12
5
14
8
3

21
  New Hampshire
451
118
71
41
19
11
6
3

71
  New Jersey
2,439
113
3
1
1
29
16
7

29
  New Mexico
710
158
128
75
34
66
38
17

128
  New York
4,595
121
4
2
1
38
21
9

38
  North Carolina
2,260
139
7
4
2
20
11
5

20
  North Dakota
407
158
66
38
17
224
141
68

224
  Ohio
3,542
125
13
8
3
43
24
11

43
  Oklahoma
1,497
305
17
10
4
35
20
9

35
  Oregon
1,194
131
37
21
9
18
10
4

37
  Pennsylvania
3,224
136
4
2
1
25
14
6

25
  Rhode Island
318
87
19
11
5
15
8
4

19
  South Carolina
1,137
145
4
2
1
6
4
2

6
  South Dakota
479
142
19
11
5
28
16
7

28
  Tennessee
1,658
126
4
2
1
15
9
4

15
  Texas
8,322
179
5
3
1
45
25
11

45
  Utah
923
135
142
82
37
105
61
27

142
  Vermont
350
114
10
6
3
60
35
16

60
  Virginia
2,044
157
2
1
1
14
8
3

14
  Washington
2,175
142
3
1
1
41
23
10

41
  West Virginia
786
159
6
3
2
11
6
3

11
  Wisconsin
1,870
155
15
9
4
20
11
5

20
  Wyoming
344
123
25
14
6
34
20
9
 
34
Public Charter



  All
3,911
255
13
7
3
13
7
3
 
13

Estimated percentages of certain population characteristics are another important feature of SASS.  For example, the percentage of traditional public schools is estimated to be 92.6 and the percentage of female principals in traditional public schools is estimated to be approximately 50.5.  To accommodate varying levels of estimated percentages in SASS, sample sizes for estimates ranging from 2.5 (rare occurrence) to 95 percentage points were evaluated against the three CV criteria.

Table 7 provides an example of the level of precision attained for estimated percentages in the 2007-08 SASS if the most stringent level of precision is applied (i.e., 15 percent CV).  The table indicates that SASS estimated percentages for both primary and high school-level public schools (traditional public and public charter combined) will have a CV no larger than 15 percent as long as the characteristic being estimated includes at least 25 percent of the target population.  

Table 7.  Estimated school-level percentages supported for the 2011-12 SASS sample allocation under the 15 percent CV criterion by public and public charter and grade level
School Type
Grade Levels
Percents Supported
Public Schools (All)
Primary/High
25 to 99

Middle / Combined
35 to 99

All
5 to 99
Public Charter Schools
All
40 to 99

Based on this analysis, the sampling frame will be partitioned into approximately 291 sampling strata for public schools.  The final number of strata will be determined after evaluating the sampling frame of eligible schools and combining strata that cannot support sampling.  The goals of the sample design are to ensure sufficient numbers for precise estimates
    • at the state level for elementary and secondary schools; 
    • at the national level for public charter schools; and
    • at the national level for middle and combined1 public schools. 

Public schools are divided into two categories to create the sampling strata: charter schools (87 strata) and traditional public schools (204 strata).

Charter schools are identified from the CCD.  The 87 charter school sampling strata are constructed by crossing the three-category school level variable (elementary, secondary, and combined) with Census region and state in the following groups:


Census Region (# Groups)

State
Census Region (# Groups)

State
Northeast (5)
Massachusetts
New Jersey
New York
Pennsylvania
All other Northeastern states
Midwest (6)
Indiana
Michigan
Minnesota
Ohio
Wisconsin
All other Midwestern states
South (8)
Delaware
District of Columbia
Florida
Georgia
Louisiana
North Carolina
Texas
All other Southern states
West (10)
Alaska
Arizona
California
Colorado
Hawaii
Nevada
New Mexico
Oregon
Utah
All other Western states

The 204 sampling strata for the traditional public schools, excluding charters schools, are defined by crossing the four-category school level variable (elementary, middle, secondary, and combined) with each of the 50 U.S. states and DC.  

Sampling – Private Schools
The same precision analysis was conducted on private schools. Table 8 presents a portion of the analysis for private schools by affiliation strata, highlighting the minimum respondent sample size needed for the 2011-12 SASS to achieve either a 15, 20, or 30 percent CV for two key estimates—the number of schools and the number of students.

Table 8. Minimum respondent sample size needed for 2011-12 SASS by percent CV for private schools by affiliation 

 
SASS 07-08
Frame
Count
SASS 07-08 Responding
Schools
Minimum Respondent Sample Size



Max %CV for Number of Schools Estimate
Max %CV for Number of Students Estimate

Max % CV
Private Schools by Affiliation


15
20
30
15
20
30

















Catholic- Parochial
3,501
295
1
0
0
12
7
3

12
Catholic- Diocesan
2,810
255
0
0
0
8
4
2

8
Catholic- Private
871
97
3
1
1
34
19
9

34
Baptist
1,835
115
7
4
2
63
36
16

63
Jewish
718
64
1
1
0
52
29
13

52
Lutheran
1,561
112
9
5
2
33
18
8

33
Seventh-Day Adventist
853
81
0
0
0
53
30
14

53
Other Religious
7,092
513
20
11
5
81
46
20

81
Nonsectarian- Regular
3,251
222
9
5
2
53
30
13

53
Nonsectarian- Special Emphasis
2,189
125
4
2
1
56
32
14

56
Nonsectarian- Special Education
1,348
89
3
2
1
30
17
8
 
30

Table 9 provides an example of the level of precision attained for estimated percentages for private schools in the 2007-08 SASS if the most stringent level of precision is applied (i.e., 15 percent CV).  The table indicates that SASS estimated percentages for elementary-level private schools will have a CV no larger than 15 percent as long as the characteristic being estimated includes at least 25 percent of the target population.  

Table 9.  Estimated school-level percentages supported for the 2011-12 SASS sample allocation under the 15 percent CV criterion by public and public charter and grade level
School Type
Grade Levels
Percents Supported
Private Schools
All
10 to 99

Elementary
25 to 99

Secondary
45 to 99

Combined 
30 to 99

Based on this analysis, the private school list sampling frame will be partitioned into 132 sampling strata defined by the interaction of the following three variables:

Stratification Variable
(# of Categories)

Variable Levels
Affiliation strata (11)
Catholic - Parochial
Catholic - Diocesan
Catholic - Private
Religious - Baptist
Religious - Jewish
Religious - Lutheran
Religious - Seventh-Day Adventist
Religious - Other
Nonsectarian - Regular
Nonsectarian - Special Emphasis
Nonsectarian - Special Education
School Type (3)
Elementary
Secondary
Combined
Census Region (4)
Northeast
Midwest
South
West

For private schools, the sample will include schools from both the cleaned PSS list frame and area frame used to identify list omissions.  All schools uniquely identified through the area frame will be included in the sample (i.e., selected with certainty).  As in previous rounds of the SASS, a relatively small number of schools have been identified through the area frame and were included with certainty.  The same procedures are expected for the 2011-12 SASS.

Sampling – Teachers Within All Schools
As with the school-level analysis, percentages of populations by important variables ranging from 2.5 (rare occurrence) to 95 percent were evaluated using the sample allocation of teachers by level of teaching experience within school type.  Table 10 summarizes the results if the most stringent level of precision is applied (i.e., 15 percent CV) to teachers within all school types—traditional public schools, public charter schools, and private schools.  This indicates that many key estimated percents will be supported by the 2011-12 SASS.  The exception is associated with rare characteristics in the population such as, for example, the percentage of teachers by non-white race/ethnicity categories by school type and grade level.

Table 10.  Estimated teacher-level percentages supported for the 2011-12 SASS sample allocation under the 15 percent CV criterion by years of experience and school type
School Type
Years of Experience
Percents Supported
Traditional Public Schools
1 year
12.5 to 99

2-3 years
10.0 to 99

4-19 years
2.5 to 99

20 or more years
2.5 to 99

School Type
Years of Experience
Percents Supported
Public Charter Schools
1 year
55 to 99

2-3 years
40 to 99

4-19 years
18 to 99

20 or more years
50 to 99
Private Schools
1 year
12.5 to 99

2-3 years
12.5 to 99

4-19 years
10 to 99

20 or more years
10 to 99

Teachers will be randomly sampled from roster information provided by each participating sampled school.  Once rostered, teachers will be randomly selected from within four sampling strata defined by level of experience:

    1. Beginning Teachers (first year of service);
    2. New Teachers (2-3 years of service); 
    3. Experienced Teachers (4-19 years of service); and
    4. Highly Experienced Teacher (20 or more years of service).

The teacher sampling rates will be constructed to meet precision requirements for estimates by level of experience and school sector (Public, Private).  This goal is met by achieving approximately 1,200 responding teachers within each of the 8 cells.

To accommodate this goal, Beginning and New Teachers in the private sector will be oversampled.  Conversely, Experienced Teachers in the public sector will be undersampled to accommodate the study budget.  The average number of teachers sampled per school will remain between three and eight with a maximum number of 20 teachers to avoid overburdening the schools.

Sampling –School Districts and Special States
Information will be collected from school districts linked to a school selected for the 2011-12 SASS.  

For Florida, Maryland, Nevada, and West Virginia, at least one school in every district in the state will be sampled.  This is necessary to reduce the size of the variances in these states.

“District-level” items are included in the School District Questionnaire, which is received by all sampled school districts, and in the Public School Questionnaire (With District Items), which is received by all charter schools and traditional public schools in single-school districts. The Public School Questionnaire (With District Items) includes both district-level and school-level items. 

Sampling – Principals Within All Schools
For all traditional public, public charter, and private schools, the principal will be included in the survey as a result of the school being selected. 

Sampling –Library Media Centers Within Public Schools
For traditional public and public charter schools, the library media center (if one exists) will be included in the survey as a result of the school being selected. 

        2.2. SURVEY WEIGHTS
Schools, principals, teachers, library media centers, and school districts will be weighted by the inverse of the probability of selection. The final weight contains adjustments for nonresponse and any other sampling or field considerations that arise after the sample has been drawn.

        2.3. RESPONSE RATES
We expect the response rates of the 2011-12 full-scale SASS to approximate those of the 2007-08 SASS or to fall slightly lower, given the long-term trend in declining response rates for federal surveys (table 11).  

Table 11.Base-weighted response rates for the 2007-08 SASS by respondent type and school sector

Unit of Observation
School Sector
Teacher
Principal
School
Library
District
  Public
84.03%
79.36%
80.36%
81.68%
87.78%
  Private
77.47%
72.18%
75.91%
†
†
† Not applicable.


    3. PROCEDURES FOR COLLECTION OF INFORMATION
The data collection methods for the 2011-12 SASS will be based on those used in the 2007-2008 SASS with modifications to the collection of the Teacher Listing Form and confirmation of the school’s eligibility as tested during the 2009-10 SASS Field Test.

Districts and schools will be mailed an advance letter between June and September 2011, depending on the size of the school district.  An internet-based Control Center will be open to schools and districts beginning in August 2011.  Through the Control Center, districts and/or schools can upload the Teacher Listing Form as well as confirm information about sampled schools, including:
 
    1. Verify the eligibility of the school;
    2. Verify the appropriate grade range of school and refer split or merged schools to Census sampling statisticians; and
    3. Identify the appropriate grade ranges to report for that school.

Questionnaires will be mailed to school districts and schools in August 2011.  Telephone follow-up will begin for districts that have not provided a Teacher Listing Form 2 weeks after the mail-out, and a reminder letter will be sent to districts that have not returned the School District Questionnaire after 3 weeks have passed (September 2011). At the school level, a package with outstanding questionnaires will be sent to schools in late September 2011, 4 weeks after the initial mail-out. . Teacher Listing Forms will be sent to schools with a nonresponding district in October 2011.  A second mail-out of questionnaires will be sent to nonresponding school districts in October 2011.  

Additional reminder letters to the school, district, and school coordinator will be sent in October and November 2011. Field staff will conduct telephone and personal visit follow-up of remaining nonresponding school districts and schools in January-April 2012.

    4. METHODS FOR MAXIMIZING RESPONSE RATES
A variety of procedures will be employed to ensure high response rates at both the level of the responding unit (i.e., sample member) and at the level of the individual survey items in each survey questionnaire.

The entire survey process, starting with securing research cooperation from key public and private school groups and individual sample members and continuing throughout the distribution and collection of individual questionnaires, is designed to increase survey response rates.  In addition, we believe that the following elements of the data collection plan, in particular, will contribute to overall success of the survey and will enhance the survey response rates.

    (1) Visible support from top-level Federal, State, and local education officials.  Without the support of high-level officials in the Department of Education, State Education Agency officials, and local school district officials representing the sampled districts, surveys of public school principals, teachers, and the library media center cannot be successfully implemented. Obtaining endorsements from these officials is a critical factor in the success of the data collection procedures. Top-level Education Department officials will need to fully support the data collection by endorsing the survey in writing and sending advance letters and notices to sampled districts' Superintendents, and individual survey participants (principals and teachers) to encourage participation.

    (2) Endorsements from key public and private school groups. The level of interest and cooperation demonstrated by key groups can often greatly influence the degree of participation of survey respondents. Endorsements are viewed as a critical factor in soliciting cooperation from state and local education officials and for obtaining high participation rates in the private sector. The SASS is seeking endorsement by the following organizations or agencies: 
 
Accelerated Christian Education
Alternative School Network
American Association of Christian Schools
American Association of Classical and Christian Schools
American Association of School Administrators 
American Association of School Librarians
American Federation of Teachers 
American Library Association
American Montessori Society
Association Montessori International
Association of Boarding Schools
Association of Christian Schools International
Association of Christian Teachers and Schools
Association of Military Colleges and Schools
Association for Supervision and Curriculum Development
Association of Waldorf Schools of North America
Christian Schools International
Council for American Private Education
Council of Chief State School Officers 
Council of the Great City Schools 
Evangelical Lutheran Church in America
Evangelical Lutheran Education Association
European Council for International Schools
Friends Council on Education
General Conference of the Seventh-Day Adventist Church
Islamic School League of America
Jesuit Secondary Education Association
Jewish Community Day School Network
Jewish Education Services of North America
Lutheran Church - Missouri Synod
National Association of Elementary School Principals 
National Association of Episcopal Schools
National Association of Independent Schools
National Association of Laboratory Schools
National Association of Private Special Education Centers
National Association of Secondary School Principals 
National Catholic Educational Association
National Christian School Association
National Coalition of Alternative Community Schools
National Coalition of Girls Schools
National Council for Private School Accreditation
National Council of Girls’ Schools
National Education Association 
National Independent Private Schools Association
North American Division of Seventh-Day Adventists
National Science Foundation
National Society for Hebrew Day Schools
Oral Roberts University Educational Fellowship
Solomon Schechter Day School Association
Southern Baptist Association of Christian Schools
Wisconsin Evangelical Lutheran Synod

    (3) Stressing the importance of the survey and the respondents' participation.  Official letters (advance notification, inside the questionnaire and follow-up) from the NCES Commissioner of Education Statistics will motivate respondents to return surveys. The additional personalization of survey materials (cover letters and survey packets with teachers' names) is expected to have positive effects on the response rates.

    (4) Minimize the survey burden on local school district and school-level authorities.  The procedures for the surveys are designed to minimize the survey burden on school districts, schools, and sampled individuals (principals, teachers, and the school librarians) and the survey instruments have been designed to be completed as quickly and easily as possible.

Good questionnaire design techniques have been employed to minimize item nonresponse. All completed questionnaires from the 2007-08 SASS have been carefully analyzed to determine which items had the highest levels of item nonresponse. This information guided NCES in reviewing the clarity of item wording, definitions, and instructions.  Items that were not considered to be effective or useful were deleted so as to streamline the questionnaires and ease the response burden. 

    (5) Seeking the recruitment of a school coordinator.  An important procedural measure for helping to maximize response rates is the plan to establish a school-based "survey coordinator" to serve as a primary point of contact for the Census Bureau.  The use of a school coordinator is expected to help keep response rates high, provide some minimal data quality checks, and simplify the follow-up process by having one point of contact.  

    5. TESTS OF PROCEDURES AND METHODS
The 2011-12 SASS was built on the experience of the previous rounds of the survey.  At the present time, we do not anticipate any testing of alternative procedures or methods in the full-scale SASS.  

    6. REVIEWING STATISTICIANS
Kerry Gruber and Kathy Chandler of NCES; Aref Dajani and Randall Parmer of the Census Bureau; Greg Strizek, contractor to the Census Bureau;  Deanna Lyter, Saida Mamedova, Yan Wang, and Jason Hill of ESSI; and  Jill Dever of RTI have all reviewed the SASS design.  Aref Dajani of the Census Bureau (301-457-1797) is the contact for further information on sample design, and Kathryn Chandler of the National Center for Education Statistics (202-502-7486) on data collection.