Question-aBy-Question Justification

ATTACHMENT D

QUESTION-BY-QUESTION JUSTIFICATION

S1-S4 Respondent Selection

To obtain a random and representative sample of persons aged 16 or older, one person in each household is selected using the most-recent birthday method. If a household in the oversample is identified as having one or more 16- to 24-year-old members, a similar random selection method will be used to choose one 16- to 24-year-old respondent. This method is widely used and is considered to be the most unobtrusive.

S5 Perceptions of Problems/Issues in the United States

The purpose of these questions is two-fold. First, the goal is to ease respondents into the format of the survey by asking them simple attitudinal questions which have no right or wrong answer, and to which most people have an opinion. Second, these questions will help NHTSA assess where U.S. residents perceive the issue of drinking and driving falling relative to other issues facing this country.

Q1 Driver Status

Respondents will be asked how frequently they usually drive a motor vehicle and those who do not usually drive will be skipped out of the parts of the interview about driving behaviors, as these questions are not appropriate for them.

Q15 Past Year Drinking

All respondents will be asked about the approximate frequency of drinking alcoholic beverages in the past year. Respondents who have not consumed alcoholic beverages in the past year will be skipped out of a large portion of the questions pertaining to drinking and driving behavior.

Q17-Q18 Type and Amount of Alcohol Typically Drunk

Question 17 will be used to classify the type of alcoholic beverage respondents consume most often. This information will allow for screening of responses to question 18 on the number of servings of different beverages consumed so that the respondents answer only for their “typical beverage.”

Q17 and Q18 (in conjunction with gender, body weight, and other information collected) allow calculation of an approximate BAC for each person’s most recent trip. This information is useful in determining the actual extent to which various groups of drivers are actually impaired.

Q19a Onset of drinking

Research has shown that age of onset of drinking impacts the engagement of drinking and can impact attitudes as well as drinking behaviors. This item will be used to segment the respondents so that different programs and campaigns can be developed to meet the needs of these different types of drinker-drivers.

Q20-Q23 Frequency and Level of Drinking in the 30-Day Period

A quantitative measure of drinking frequency within a typical 30-day period will be obtained in question 20. This information will be used to categorize frequent and infrequent drinkers in the analysis, and to permit analyses looking at the question of whether frequent and infrequent drinkers contribute differentially to the drinking-driving problem. Question 23 will collect information on the extent to which the respondent engages in heavy or “binge” drinking in a typical month. It will allow NHTSA to better evaluate the perceptions and behaviors of heavy drinkers as compared to moderate drinkers so that programs can be designed to be more effective in reducing drinking-driving episodes among each group.

Q26-Q29 Drinker Classification

The four perceptual questions about the respondent’s own drinking behaviors occurring in this series make up an index to identify problem drinkers. Such classifications are useful to identify the proportion of the population who report symptoms of problem drinking to assess if this subsegment is growing or decreasing over time. In addition, the attitudes and behaviors of problem drinkers (who are over-involved in drinking driving trips) are important to understand when developing countermeasures appropriate to this group.

Q31 Perceptions of Self Impairment

Question 31 will assess how much alcohol of the respondents’ “typical” alcoholic beverage they feel they can drink and still drive. While the extent of impairment varies with BAC for different people and depends on certain circumstances, it is important to understand how a person’s perception of self-impairment differs from what would be expected at specified consumption amounts. This information will help NHTSA determine the kinds of public education needed to help people understand the true extent of impairment that can result from different levels of drinking. Developing strong public attitudes against alcohol-impaired driving will not be possible if the public holds inaccurate perceptions of the effects of alcohol on driving.

Q33-Q36 Frequency of Drinking and Driving

Question 33 measures the prevalence of driving a motor vehicle within two hours of drinking alcohol. Question 34 quantifies the frequency of this behavior in the past year. These items provide objective, rather than subjective (“Do you think you were impaired?”), measures of drinking and driving prevalence, and, as such, they provide the keys to estimating the magnitude of the problem and tracking its direction over time. Question 33 is the key measure NHTSA uses to trend the prevalence in drinking-driving behavior over time (this measure has shown slight declines since 1991).

Questions 35 and 36 will quantify drinking and driving behavior in the shorter and more recent time period of 30 days. This measure of more recent behavior more accurately represents the behavior of interest by those who engage in it quite frequently. This measure is key to the analysis of estimating total annual drinking-driving trips by various subgroups.

Evaluation of drinking and driving during the past 30 days will also be compared to total reported drinking in that same time frame as measured in Question 20 to obtain a measure of the relative frequency of driving after drinking. A person who drives after every drinking occasion has a very different problem (and needs to be treated differently) than one who drinks frequently, but refrains from driving on almost all of those occasions.

Q37-Q44a Most Recent Experience of Drinking and Driving

Questions 37 through 44a ask respondents to think about the most recent occasion when they drove within two hours of consuming alcoholic beverages. Measuring behaviors for the most recent occasion will provide for the greatest accuracy in representing specifics of each person’s drinking and driving behaviors.

Question 37 identifies the place where the respondent drank on his or her most recent drinking occasion. NHTSA uses this information along with other survey data (such as amount consumed and distance driven) to compare risks associated with different drinking origins. This will allow NHTSA to develop specific strategies and safety messages for both drinkers and host/proprietors about unsafe drinking and driving practices.

Question 38 measures the actual number of drinks consumed on this most recent drinking and driving occasion, while Q39 assesses over what time frame these drinks were consumed. Question 41 identifies the length of time between the respondent’s last drink and his or her departure. Data from this series of questions will be used (along with the respondent’s sex and body weight) to estimate BAC level at the time of departure. This information allows for an estimate of the prevalence of impaired driving at different BAC levels. This information is not usually acquired, and will be extremely useful in identifying sub-segments of the population in most need of countermeasures. Question 43a asks if the respondent wore a seat belt on this drinking and driving occasion. This will allow NHTSA to assess the relationship between BAC level and belt use.

Question 44 asks the total number of people in the vehicle during the drinking-driving episode and will allow for estimates of total at-risk persons. Question 44a will provide data on the number of youths at-risk.

Q49 Perception of Self Impairment on Most Recent Occasion

The final question in the most recent occasion series asks the respondents if they believed that they were over or under the legal limit on that occasion. This item will help gauge what type of drinker-drivers are most likely to drive knowingly impaired and to estimate the prevalence of such behavior among the general population. Also, the calculated BAC levels will be compared to reported self impairment to assess the relative accuracy of self-reported impairment.

Q52 Frequency of Driving While Over the Legal Limit

This question will provide an overall estimate of the frequency of impaired driving over a twelve month period. NHTSA will use these data to identify frequent impaired drivers.

Q53 Number of Drinks Consumed to Put Self Over the Legal Limit

Question 53 provides data on public perceptions of how much alcohol it takes to reach the legal limit. This information is needed because of the importance of legal limits in controlling alcohol-impaired driving.

Q54-Q56 Avoidance of Drinking and Driving

Given that alcohol will continue to be widely used by the general driving public, it follows that efforts must be directed at finding ways to separate the activity of drinking from the activity of driving by getting people to avoid driving after drinking. Question 54 will provide estimates of the proportion of the potentially at-risk (for alcohol-impaired driving) drivers who have avoided driving at least once in the past year because they felt they were too impaired to drive. Trending of this question offers one gauge of the effectiveness of initiatives to encourage people not to drive after drinking.

Question 56 assesses what actions the respondent took to deliberately avoid driving after drinking and is useful in developing countermeasures that are acceptable to the general population and to various subgroups.

Q57 Riding with Impaired Driver in the Past Year

Passengers riding in motor vehicles operated by alcohol-impaired drivers are at as much risk as drivers who are impaired themselves. Question 57 will assess the proportion of respondents who have ridden with a driver they thought may have been too impaired to drive safely.

Q61-Q66 Designated Drivers

Designated driver programs have the potential for reducing alcohol-impaired driving and these programs enjoy wide public support. Questions 61 through 66 measure experience with designated driver situations, personal experience as a designated driver, and acceptance of alcohol use by the designated driver.

Question 61 quantifies episodes of riding with designated drivers. Question 62 will quantify the extent to which designated drivers limit their consumption of alcohol. The timing of the decision to have a designated driver in Question 63 will provide further input into how designated driver programs should be targeted and help determine if such programs and campaigns have been effective.

Quantification of the respondent as a designated driver during the past year is measured in Question 64b through Question 65. Questions 65 and 65a collect information on how the designated driver activity was implemented, which will inform NHTSA as to whether the activity is being performed in an effective manner.

Question 66 will measure whether the public understands that a designated driver shouldn’t consume any alcohol at all. If the public does not understand this, it would point to the need for a renewed educational intervention.

Q86-Q95 Personal Responsibility of Serving Alcohol to Adults and Minors

One approach to controlling alcohol-impaired driving is to encourage personal responsibility when serving alcohol. It is important for NHTSA to measure the effectiveness of its communications on the importance of the responsible serving of alcohol and to determine what information the public lacks.

The Question 86 series assesses respondents’ self-reported behavior regarding serving alcohol at social events and their understanding about signs of impairment. This includes asking about serving alcohol to underage drinkers. Question 87 asks how concerned they were about guests driving away impaired. The information will be used to determine if hosts are sufficiently knowledgeable and concerned about the risks that this activity poses, and whether they have a sense of responsibility in not making alcohol available to teens.

Question 95 will assess the use of various methods to prevent guests from driving home impaired.

Q96-Q102 Intervention

Another potentially effective means to reduce alcohol-impaired driving is to encourage intervention by friends before an impaired driver gets behind the wheel. This is the premise of the “Friends Don't Let Friends Drive Drunk” campaign.

Question 96 asks the respondents if they have been in a situation where a friend was too impaired to drive. If yes, the survey explores if the respondent took action, what action was taken, and whether it was effective. Question 100 will provide data on the proportion of the population that has intervened to prevent a friend from driving while impaired. Question 102 measures the effectiveness of interventions identified in Question 101 that were used to prevent impaired driving by a friend.

NHTSA will use the information in this series to measure efforts to encourage intervention by friends and to trend the acceptance and performance of the behavior over time.

Q102a Social Pressure

This series of questions measures situations of peer pressure which may have led respondents to drink when they otherwise may not have. Understanding what types of situations may arise can help NHTSA design messages and countermeasures that address the most frequent occurrences.

Q103-Q114 Perceptions of the Scope of Drinking and Driving and Social Punishment

It is important for NHTSA to trend public perceptions of the social impact of and punishment for drinking and driving. Trending of these data informs NHTSA of progress and changes it needs to make in its public awareness efforts and support of countermeasures and laws. The more that the general public sees the safety of themselves and their families threatened by impaired drivers, the more supportive they are likely to be of efforts to control these behaviors.

Questions 103 through 116b measure the public’s perceptions of the impact of drinker-drivers and the social punishment they feel is due to those convicted for violating drinking and driving laws. Question 103 will assess the perceived threat to the personal safety of the respondent and his or her family from someone who drinks and drives.

The Question 105 series will assess the public’s perceptions of how likely drinking and driving will lead to negative outcomes. Will the person be likely to be stopped by police or be involved in a crash? If stopped by police, will the person be arrested and convicted? The answers to these questions will help NHTSA assess the perceived threat and deterrent value of these possibilities.

A key component of this study is to gauge the extent to which public opinion and tolerance regarding drinking and driving has changed since the inception of the study. Understanding the public’s views on people who drink and drive (Question 104-A) and laws mandating zero tolerance for alcohol (Question 104c) will aid NHTSA in designing programs that can be accepted by a large proportion of the public and that are likely to be effective.

Questions 113 and 114 ask about personal experience of arrest for drinking and driving violations in the past two years. Besides providing a measure of enforcement activity, responses to these questions will be compared to self-reported drinking and driving behavior to assess the relationship of behavior to consequences.

Q116 Severity of Enforcement

Understanding the level of penalty that is acceptable to the public for violating drinking and driving laws can help States and municipalities determine if there is public support for strengthening the drinking and driving laws.

Q120 -Q122b Sobriety Check Points and Other Enforcement Interventions

Questions 120 through 122 will measure exposure to, and support for, sobriety checkpoints as a means of deterring impaired driving. NHTSA will then be able to provide data on public support of checkpoints to the States and various communities. Questions 122a and 122b will assess the extent to which the public is aware of other approaches law enforcement takes to enforce drinking and driving laws. This will indicate whether those interventions are being implemented in a way that makes an impression on their targeted audiences.

Q123-Q126 Blood Alcohol Concentration Limits

This question series examines the public’s familiarity with the concept of Blood Alcohol Concentration (BAC), their knowledge of the BAC limits specified in their State laws, and their perceptions of how much alcohol they could drink before reaching the BAC limit. The purpose of this section is to determine if the public has an accurate or inaccurate perception of how much they could drink before being in violation of their State drinking and driving laws.

Q131a-Q133 Crash/Injury Exposure

As the key focus of all of the programs supported by NHTSA is to reduce alcohol-related crashes and fatalities, the data from Questions 131a-133 allow an estimate of the effectiveness of their programs in accomplishing the goal. These questions assess involvement in alcohol-related motor vehicle crashes in the past two years.

Questions 131a and Q133 determine whether the respondent has had any alcohol-related motor vehicle crashes within the past two years and, if so, whether anyone was injured in the crash(es).

Q139 Perceptions of Effectiveness of Program Initiatives

This series of questions was added in 1999 by the urging of NIAAA to measure perceptions of the effectiveness of strategies that are believed to affect and reduce drinking and driving. The series covers the main areas of the environmental and enforcement efforts. Since public acceptance and “buy in” of programs and strategies is one of the underlying factors driving adherence, these questions will be crucial to NHTSA and NIAAA in understanding which strategies need additional public messages, which are widely accepted as effective and which may need to be redirected to be perceived as effective by the community.

Q139a-140a-d Penalties for Underage Offenders and Repeat Offenders

Youth is a population of particular concern with respect to drinking and driving. Question 139a obtains information on the perceived effectiveness of different interventions directed toward this population. There is also the issue of how to deter first time offenders from continuing the problem behavior. Question 140 asks respondents the appropriateness of different interventions for first time offenders. Both question series thereby measure public support for different intervention approaches as well as the acceptability of targeting these populations.

D1-D11 Demographic Variables

The demographic data collected in this study serve two primary purposes. The first is to monitor and evaluate the representativeness of the surveyed population as compared to the total target population for the United States. This allows the final data to be weighted by number of adults in the household, age, gender, race, education, and number of household telephone lines, and to correct any sampling disproportionalities by ensuring that the demographic profile of the surveyed population reflects the known population parameters of persons aged 16 and older living in the United States.

The second purpose of collecting the demographic data is to be able to examine differences in attitudes and behaviors regarding drinking and driving among different demographic sub groups of the population. This information will allow NHTSA to target education and programs to the highest at-risk groups.

Question D1 will record the respondent's age.

Question D1b assesses the number of children in the household.

Question D2 will record whether the respondent is employed.

Question D3 will record the respondent's level of education.

Question D4 will record the respondent's marital status.

Questions D5-D6 will record the respondent's ethnic background and race.

Question D7a will assess if the respondent is a licensed driver.

Question D8 will record respondent household income for 2006.

D9 Respondent Weight

Weight is needed to estimate BAC levels of people who drink and drive. Among other things, this will permit a determination of the impairment levels that respondents associate with having an effect on their driving skills, and becoming too dangerous to drive.

D10 Sex

Question D10 will record the sex of the respondent. The interviewer codes this question unless it is unclear at the end of the interview.

D11a-D11 Multiple Telephone Numbers

These questions are used to weight households on the basis of unequal opportunity for selection based on multiple telephone numbers for a single household.