A Strategic Approach to Transforming Traffic Safety Culture to Reduce Deaths and Injuries (2018)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

B-1 Appendix B – Guidelines for Design of Surveys (Task 2 Report) Recommendations for question design with respect to each component of traffic safety culture are provided below along with example questions to examine the specific behavior of speeding among the specific group of teen drivers in high school. Formative research is required prior to creating questions. This research identifies details about the specific behavior (such as words used to describe the behavior) as well as other components of the model (such as attitudes, behavioral beliefs, etc.). Recommendations for Best Practice in Question Design for Measuring Traffic Safety Culture Measuring Behavior When creating questions to measure the behavior, we recommend addressing the following questions first: 1. What is the action performed (being as specific as possible)? 2. Who engages in the action and to whom is the action directed (if appropriate)? 3. In what context does the action take place? 4. When is the action performed? For example, suppose we want to address the issue of speeding: What is the action performed? In this case, we will define speeding as exceeding the posted speed limit by 10 mph or more. Who engages in the action and to whom is the action directed? We want to examine speeding behaviors among drivers in high school. In what context does the action take place? We will examine speeding on roads with posted speed limits of 55 mph and above. When is the action performed? We will not limit the behavior by a certain time. (We could explore day versus night or weekday versus weekend, etc.) Answering these questions provides the specificity needed to obtain accurate information (Fishbein & Ajzen, 2009). Furthermore, it is important to keep the responses

B-2 to these questions in mind throughout the design of the survey to make sure that all constructs align with the behavior as described as much as possible. Next, questions are created to measure the behavior (as described in answering the questions about action, target, context, and time – see above). Whenever possible, simple language is used to allow for comprehension by a wide range of education levels. Answer choices are created that provide a wide range of possible answers distributed across a 7- point scale (the even distribution across a 7-point scale supports better analyses) (Fishbein & Ajzen, 2009). Example In the past 30 days, how often have you exceeded the speed limit by more than 10 mph when driving on roads with posted speed limits of 55 mph or more? Never, Rarely, Occasionally, About half the time, Frequently, Usually, Always Results are scored from 1 (never) to 7 (always). Measuring Intention Intention is the deliberate decision to commit a behavior in an anticipated situation. The same language used to describe the behavior in measuring the actual behavior should be used in measuring intention. Language such as “I will …,” “I intend…,” or “I expect…” is often used. Example In the next 30 days, how often do you intend to exceed the speed limit by more than 10 mph when driving on roads with posted speed limits of 55 mph or more? Never, Rarely, Occasionally, About half the time, Frequently, Usually, Always Responses are scored from 1 (never) to 7 (always). Their scoring aligns with the scoring used to measure the speeding behavior. Measuring Willingness Willingness is the predisposition to commit a behavior if an unexpected situation arises. For a given circumstance, various levels of engagement are listed, and respondents indicate how willing they are to engage in the behavior in that circumstance. For example, various levels of engagement for drinking might be having one drink with a meal; have 2 – 3 drinks;

B-3 drinking to the point of losing control; drinking to the point of passing out. However, not all behaviors have various levels of engagement. For example, there are not really varying levels of wearing a seat belt. The following is an example of measuring willingness for speeding. Various levels of engagement are represented by increasing speed. Example Imagine you are driving on a highway with a posted speed limit of 55 mph. There are not many vehicles on the road, and two friends are in the car with you. How willing would you be to try the following? Drive at 65 mph Not at all willing | | | | | | | | Very willing Drive at 70 mph Not at all willing | | | | | | | | Very willing Drive at 75 mph Not at all willing | | | | | | | | Very willing Drive at 80 mph Not at all willing | | | | | | | | Very willing These responses are scored from 1 (not at all willing) to 7 (very willing). They are averaged to create a scale for willingness. Measuring Attitudes Attitudes are subjective evaluations of an object or behavior in terms of emotional reaction (e.g., “Speeding is exciting”) and perceived utility (e.g., “Seat belts are useless). Semantic differentials are an effective way to measure attitudes (Fishbein & Ajzen, 2009). A semantic differential offers word pairs which are opposite adjectives, and respondents indicate to what degree they believe one word or the other describes the behavior. Adjectives seek to address both instrumental outcomes (such as “dangerous / safe”) as well as experiential outcomes (such as “pleasant / unpleasant”). For the speeding example, we use the four adjective pairs developed by Conner et al. (2007). Example Please indicate how you feel about each of the following statements.

B-4 Exceeding the speed limit by more than 10 mph when driving on roads with posted speed limits of 55 mph or more would be: Very dangerous Very irresponsible | | | | | | __ | | | | | | | | __ | | Very safe Very responsible Very stupid | | | | | | __ | | Very sensible Very unpleasant | | | | | | __ | | Very pleasant Responses are scored from -3 (negative attitude toward speeding) to +3 (positive attitude toward speeding). Responses are averaged to form an overall attitude scale. Measuring Behavioral Beliefs Behavioral beliefs are expectations about the physical and social consequences of a behavior (e.g., “If I speed, I will likely get an expensive fine,” “If I drink and drive, my friends will exclude me”). Behavioral beliefs predict attitudes. Measuring behavioral beliefs increases our understanding of the sources of attitudes. We recommend measuring behavioral beliefs using a pair of questions for each potential belief: one for behavioral strength and one for outcome evaluation (Fishbein & Ajzen, 2009). A behavioral strength is measured with a conditional statement such as “If I don’t answer my cell phone while driving, I feel my spouse may be upset with me.” An outcome evaluation assesses how important the associated outcome is: “Not upsetting my spouse is very important to me.” Behavioral beliefs should address both instrumental outcomes (i.e., a means to an outcome) and experiential or emotional outcomes. Using conditional phrases encourages responses regardless of whether the individual engages in the behavior or not (Hampson et al., 2001). Language that connects with feelings as opposed to rational thought also has been found to be more predictive of actual intentions and behaviors (Weinstein et al., 2007). Example If I exceed the speed limit by more than 10 mph, I feel that I will get to my destination much quicker. Strongly agree, Agree, Somewhat agree, Neither, Somewhat disagree, Disagree, Strongly disagree Getting to my destination much quicker is very important to me.

B-5 Strongly agree, Agree, Somewhat agree, Neither, Somewhat disagree, Disagree, Strongly disagree These responses are scored from 1 (strongly disagree) to 7 (strongly agree). Their product is positive because they align with a positive attitude about speeding. If I exceed the speed limit by more than 10 mph, I feel that I am more likely to have a crash / accident. Strongly agree, Agree, Somewhat agree, Neither, Somewhat disagree, Disagree, Strongly disagree Not having a crash / accident is very important to me. Strongly agree, Agree, Somewhat agree, Neither, Somewhat disagree, Disagree, Strongly disagree These responses are scored from 1 (strongly disagree) to 7 (strongly agree). Their product is made negative because they align with a negative attitude about speeding. If I exceed the speed limit by more than 10 mph, I feel that I may place others at greater risk. Strongly agree, Agree, Somewhat agree, Neither, Somewhat disagree, Disagree, Strongly disagree Not placing others at greater risk is very important to me. Strongly agree, Agree, Somewhat agree, Neither, Somewhat disagree, Disagree, Strongly disagree These responses are scored from 1 (strongly disagree) to 7 (strongly agree). Their product is made negative because they align with a negative attitude about speeding. If I exceed the speed limit by more than 10 mph, I feel that driving is more fun. Strongly agree, Agree, Somewhat agree, Neither, Somewhat disagree, Disagree, Strongly disagree Making driving more fun is very important to me. Strongly agree, Agree, Somewhat agree, Neither, Somewhat disagree, Disagree, Strongly disagree These responses are scored from 1 (strongly disagree) to 7 (strongly agree). Their product is positive because they align with a positive attitude about speeding. All the products are averaged to form a scale for behavioral beliefs.

B-6 Measuring Prototype Image Prototype image is the stereotype of people perceived to typically engage in the behavior (e.g., “People who speed are cool”). Measuring an individual’s image beliefs involves listing a series of adjectives and asking respondents to indicate how well each adjective aligns with their image of the typical person who engages in the behavior. The process is repeated for a typical person who does not engage in the behavior. The same adjectives are used for both images. Example Think about the “typical” person your age who regularly speeds while driving (that is, exceeds the posted speed limit by 10 mph or more). How much does each of the following words fit the image of this “typical” person? Attractive Stupid* Cool Dangerous* Skilled Careless* Exciting Immature* Does not fit at all | | | | | | | | fits very well Does not fit at all | | | | | | | | fits very well Does not fit at all | | | | | | | | fits very well Does not fit at all | | | | | | | | fits very well Does not fit at all | | | | | | | | fits very well Does not fit at all | | | | | | | | fits very well Does not fit at all | | | | | | | | fits very well Does not fit at all | | | | | | | | fits very well These responses are scored from 1 (does not fit at all) to 7 (fits very well). Words noted with an asterisk (*) are reversed coded: 7 (does not fit at all) to 1 (fits very well) because they are negative adjectives. All responses are averaged to form a scale for the prototypical image of someone who engaged in the behavior.

B-7 Now think about the “typical” person your age who regularly drives within 10 mph of the posted speed limit (that is, they do NOT speed). How much does each of the following words fit the image of this “typical” person? Attractive Stupid* Cool Dangerous* Skilled Careless* Exciting Immature* Does not fit at all | | | | | | | | fits very well Does not fit at all | | | | | | | | fits very well Does not fit at all | | | | | | | | fits very well Does not fit at all | | | | | | | | fits very well Does not fit at all | | | | | | | | fits very well Does not fit at all | | | | | | | | fits very well Does not fit at all | | | | | | | | fits very well Does not fit at all | | | | | | | | fits very well These responses are scored from -1 (does not fit at all) to -7 (fits very well). Words noted with an asterisk (*) are reversed coded: -7 (does not fit at all) to -1 (fits very well) because they are negative adjectives. Negative numbers are used because this image is associated with NOT engaging in the behavior. All responses are averaged to form a scale for the prototypical image of someone who does not engage in the behavior. Measuring Perceived Norms Perceived norms are the behavior believed to be common and expected in a given context (e.g., wearing a seat belt when driving with parents). Both injunctive and descriptive perceived norms should be measured as both contribute to perceived social pressure to engage (or not engage) in a behavior (Fishbein & Ajzen, 2009). Following the reasoned action approach, a single question using a general referent group is sufficient for each.

B-8 Injunctive perceived norms address what we think others believe is appropriate behavior and often use language such as “expect me” or “should do.” Descriptive perceived norms address what we think most actually do. The description of the behaviors for the perceived descriptive norms should align with the language used to measure the behaviors. Example – Perceived Injunctive Norm Please indicate how much you agree or disagree with the following statement. “People who are important to me expect me never to exceed the posted speed limit by more than 10 mph when driving on roads with posted speed limits of 55 mph or more.” Strongly agree, Agree, Somewhat agree, Neither, Somewhat disagree, Disagree, Strongly disagree These responses are scored from -1 (strongly disagree) to -7 (strongly agree). The negative sign is used because this injunctive norm addresses NOT engaging in speeding behavior. Example – Perceived Descriptive Norms In the past 30 days, how often have most people like you exceeded the speed limit by more than 10 mph when driving on roads with posted speed limits of 55 mph or more? Never, Rarely, Occasionally, About half the time, Frequently, Usually, Always These responses are scored from 1 (never) to 7 (always). Their scoring aligns with the scoring used to measure the speeding behavior. The responses to injunctive norms and perceived norms may be averaged to form a single scale for perceived norms. Measuring Normative Beliefs Normative beliefs include beliefs about what behaviors are most common in a specific group (e.g., “All my friends speed”) and what people in a specific group expect (e.g., “My parents expect me to wear a seat belt”). Normative beliefs predict perceived norms. While perceived norms assess normative beliefs about a non-specific group (“those important to me”), normative beliefs are assessed about a specific group. Different referent groups (e.g., “my parents,” “my friends,” “adults in my community”) create different levels of social pressure. Measuring the relative influence of normative beliefs for specific groups enhances our overall understanding.

B-9 Both injunctive and descriptive normative beliefs should be measured. We recommend measuring injunctive normative beliefs using a pair of questions for each potential belief: one for belief strength and one for motivation to comply (Fishbein & Ajzen, 2009). Belief strength is measured using a statement such as “My close friends think I should never drive after drinking.” Motivation to comply is measured using a statement such as “In general, I want to do what my close friends think I should do.” These questions are repeated for different referent groups. Descriptive normative beliefs are measured with one question for each referent group. Each question measures the respondent’s perception of engagement in the behavior by the specific referent group (for example, “How often do most of your close friends drive after drinking?”). We recommend using the language “How often do most…” The behavior (and answer choices) should be described using the same language as is used in the questions measuring the behavior. Example “My parents / guardians think I should never exceed the posted speed limit by more than 10 mph when driving on roads with posted speed limits of 55 mph or more.” Strongly agree, Agree, Somewhat agree, Neither, Somewhat disagree, Disagree, Strongly disagree “In general, I want to do what my parents / guardians think I should do.” Strongly agree, Agree, Somewhat agree, Neither, Somewhat disagree, Disagree, Strongly disagree “My friends think I should never exceed the posted speed limit by more than 10 mph when driving on roads with posted speed limits of 55 mph or more.” Strongly agree, Agree, Somewhat agree, Neither, Somewhat disagree, Disagree, Strongly disagree “In general, I want to do what my friends think I should do.” Strongly agree, Agree, Somewhat agree, Neither, Somewhat disagree, Disagree, Strongly disagree “Teachers and other adults at school whom I respect think I should never exceed the posted speed limit by more than 10 mph when driving on roads with posted speed limits of 55 mph or more.” Strongly agree, Agree, Somewhat agree, Neither, Somewhat disagree, Disagree, Strongly disagree “In general, I want to do what my teachers and other adults at school whom I respect think I should do.” Strongly agree, Agree, Somewhat agree, Neither, Somewhat disagree, Disagree, Strongly disagree

B-10 These responses are scored from 1 (strongly disagree) to 7 (strongly agree). The results of each pair of questions are multiplied together and given a negative sign. The negative sign is used because these injunctive norms address NOT engaging in speeding behavior. The products are averaged to form a sub-scale for normative injunctive beliefs. In the past 30 days, how often have your parents / guardians exceeded the speed limit by more than 10 mph when driving on roads with posted speed limits of 55 mph or more? Never, Rarely, Occasionally, About half the time, Frequently, Usually, Always In the past 30 days, how often have most students in your school who can drive exceeded the speed limit by more than 10 mph when driving on roads with posted speed limits of 55 mph or more? Never, Rarely, Occasionally, About half the time, Frequently, Usually, Always In the past 30 days, how often have most adults in your community exceeded the speed limit by more than 10 mph when driving on roads with posted speed limits of 55 mph or more? Never, Rarely, Occasionally, About half the time, Frequently, Usually, Always These responses are scored from 1 (never) to 7 (always). Their scoring aligns with the scoring used to measure the speeding behavior. The responses are averaged to form a sub- scale for normative descriptive beliefs. Measuring Perceived Control Perceived control is the perception of our ability to determine own behaviors (e.g., “I can choose my own speed in traffic”). In the case of some risky behaviors, it may be more appropriate to ask about perceived control to not engage in the behavior. For example, in the case of drinking and driving, we would seek to understand to what degree individuals believe they are capable of not driving after drinking. For some behaviors, it may be appropriate to measure perceived control in several different situations. For example in the case of seat belt use, different situations could include when alone, when with friends, and when driving on the highway. Example How much do you agree or disagree with the following statements? “If I really wanted to, I could drive within 10 mph of the posted speed limit when driving on roads with posted speed limits of 55 mph or more.” Strongly agree, Agree, Somewhat agree, Neither, Somewhat disagree, Disagree, Strongly disagree

B-11 These responses are scored from -1 (strongly disagree) to -7 (strongly agree). The negative sign is used because agreement aligns with NOT engaging in speeding behavior. Measuring Control Beliefs Control beliefs are beliefs about our ability to engage or not engage in the behavior based on factors that are either internal or external to oneself (e.g., “Crashes are determined by fate,” “It is up to me whether I speed or not even if everyone around me is”). Perceived control measures an individual’s general sense of being able to control a behavior. A controlling belief describes the influence of a specific factor or scenario that may impact an individual’s sense of being in control of the specific behavior. Control beliefs predict perceived control. Measuring control beliefs provides a better understanding of the sources of perceived control. We recommend measuring control beliefs using a pair of questions for each potential belief: one to assess the likelihood of the control factor or scenario and one for the power of this factor on the sense of control (Fishbein & Ajzen, 2009). The likelihood of the control factor can be assessed using a question such as “How likely is it that your spouse will call you on your cell phone when you are driving?” To assess the power of the factor, a question about whether this factor would increase (or decrease) control is asked such as “Having my spouse call me on my cell phone while driving makes it more likely I will answer.” The product of these two responses is an indication of the relative weight of this control factor relative to other control factors measured. Example “How likely is it that your friends will pressure you to speed when they are in the car with you?” Extremely likely, Likely, Somewhat likely, Neutral, Somewhat unlikely, Unlikely, Extremely unlikely “Not having friends with me in the car pressuring me to speed would make it easier not to speed.” Strongly agree, Agree, Somewhat agree, Neither, Somewhat disagree, Disagree, Strongly disagree “How likely is it that vehicles around you will be speeding when you drive?” Extremely likely, Likely, Somewhat likely, Neutral, Somewhat unlikely, Unlikely, Extremely unlikely “Not having vehicles around me speeding makes it easier not to speed.”

B-12 Strongly agree, Agree, Somewhat agree, Neither, Somewhat disagree, Disagree, Strongly disagree “How likely is it that you regularly notice the posted speed limit and your own speed?” Extremely likely, Likely, Somewhat likely, Neutral, Somewhat unlikely, Unlikely, Extremely unlikely “Regularly knowing the posted speed limit and my own speed makes it easier not to speed.” Strongly agree, Agree, Somewhat agree, Neither, Somewhat disagree, Disagree, Strongly disagree These responses are scored from 1 (extremely unlikely, strongly disagree) to 7 (extremely likely, strongly agree). The results of each pair of questions are multiplied together. The product of the last pair is assigned a negative value because increased value corresponds to not speeding. The products are averaged to form a scale of control beliefs. Measuring Values Values are ideals to which we aspire that define the goals for our behavioral choices and direct the formation of our belief systems (e.g., “I must protect my family,” “I desire a life without stress”). We recommend using the most-least rank rate method (McCarty & Shrum, 2000) described in Appendix A applied to the Short Schwartz Value Survey (SSVS) (Lindeman & Verkasalo, 2005). This approach measures a range of values with a limited number of questions. Because it is recognized that values are held independent of a specific behavior (Spates, 1983), the behavior addressed in the survey is not mentioned in any of the questions about values. While the language in these questions is written at a higher level, we recommend using validated scales instead of changing the language. Future research could explore adapting these scales. Example Please review the following list of values and identify the one that is MOST IMPORTANT to you. Please write it down here: Now review the list again and identify the one that is LEAST IMPORTANT to you. Please write it down here: . Please rate how important each of the following are to you. 1. Universalism (that is, broad mindedness, beauty of nature and arts, social justice, a world at peace, equality, wisdom, unity with nature, environmental protection) Opposed to my principles Not important (2) (3) (4) (5) (6) (7) Of supreme importance

B-13 (0) (1) (8)          2. Benevolence (that is, helpfulness, honesty, forgiveness, loyalty, responsibility) Opposed to my principles (0) Not important (1) (2) (3) (4) (5) (6) (7) Of supreme importance (8)          3. Conformity (that is, obedience, honoring parents and elders, self-discipline, politeness) Opposed to my principles (0) Not important (1) (2) (3) (4) (5) (6) (7) Of supreme importance (8)          4. Tradition (that is, respect for tradition, humbleness, accepting one’s position in life, devotion, modesty) Opposed to my principles (0) Not important (1) (2) (3) (4) (5) (6) (7) Of supreme importance (8)          5. Security (that is, national security, family security, social order, cleanliness, reciprocation of favors) Opposed to my principles (0) Not important (1) (2) (3) (4) (5) (6) (7) Of supreme importance (8)          6. Power (that is, social power, authority, wealth) Opposed to my principles (0) Not important (1) (2) (3) (4) (5) (6) (7) Of supreme importance (8)          7. Achievement (that is, success, capability, ambition, influence on people and events) Opposed to my principles (0) Not important (1) (2) (3) (4) (5) (6) (7) Of supreme importance (8)          8. Hedonism (that is, gratification of desires, enjoyment in life, self-indulgence) Opposed to my principles (0) Not important (1) (2) (3) (4) (5) (6) (7) Of supreme importance (8)         

B-14 9. Stimulation (that is, daring, a varied and challenging life, an exciting life) Opposed to my principles (0) Not important (1) (2) (3) (4) (5) (6) (7) Of supreme importance (8)          10. Self-Direction (that is, creativity, freedom, curiosity, independence, choosing one’s own goals) Opposed to my principles (0) Not important (1) (2) (3) (4) (5) (6) (7) Of supreme importance (8)         