PHE 505 Module 7 Survey Assignment Example

Why Not a Helmet? Designing and Pretesting a Survey of Shared Scooter Riders

[Student Name]

Southern New Hampshire University

PHE 505: Research Methods in Public Health

Module Seven Survey Assignment

[Instructor Name]

[Date]

The organization, setting and figures below are a composite written as a model document. No real employer, client, colleague or patient is described.

Why Not a Helmet? Designing and Pretesting a Survey of Shared Scooter Riders

Purpose

Hospital records can show whether helmet nonuse is linked to head injury, but they cannot explain why so few riders wear helmets. A short survey of riders can. Its results would help the city and the scooter companies decide whether to supply helmets, run campaigns or require helmets for some riders. The survey is a companion to the record study, not part of its main analysis.

Constructs

The survey measures five constructs: how often riders wear a helmet, why they do not, how risky they think scooter riding is, whether they ride after drinking and whether they would use a helmet if one came with the scooter. A few demographic items allow comparison with the injured patients in the record study.

Each construct links to a decision the city faces.

How the Items Were Written

Dillman et al. (2014) describe the tailored design method, which aims to make answering easy and worthwhile for respondents. Several of their principles guided the wording. Each question asks about one thing only. Response options are balanced and do not overlap. Time frames are specific, such as the past 30 days, rather than usually. Sensitive items, such as riding after drinking, come late, after trust is established, and are worded without judgment. The survey is short enough to finish while waiting for a bus, about three minutes.

The Items

Table 1 lists the twelve items after pretesting.

Table 1. Survey items

Note. Items 6 and 8 were revised after pretesting.

Pretesting

The draft was tested in cognitive interviews with six riders recruited near a scooter parking area on campus. Each read the questions aloud and explained what they thought each meant and how they chose an answer. Three findings led to changes. First, the original drinking item, asking whether riders had ridden under the influence, was read as asking about legal intoxication, so it was rewritten with a time window. Second, riders interpreted a helmet as including bicycle helmets they owned but rarely carried, which confirmed that reason options should include did not have one with me. Third, a price-free question about buying a helmet produced near-universal yes answers, so a price was added. Pretesting took about two hours and saved the survey from three misleading items.

Sampling and Mode

The target population is adults who rode a shared scooter in the past 30 days. A probability sample is not possible because no list of riders is available to the study. The survey will use intercept sampling at the eight busiest scooter parking areas, at set times across weekdays, weekends and evenings, with a QR code linking to a mobile survey and paper copies for those who prefer them. Trivedi et al. (2019) combined hospital records with observation of riders at intersections; a similar observation tally at each site will estimate how many riders passed and how many took the survey, giving a rough response rate. A target of 400 completed surveys would estimate the share who never wear a helmet within about five percentage points.

Ethics

The survey is anonymous and collects no names, emails or locations. A short statement at the start explains the purpose, that participation is voluntary and that any question can be skipped. Because it involves minimal risk and no identifiers, it would likely qualify for exempt review, but the decision belongs to the review board.

Reliability and Validity

Creswell and Creswell (2018) note that researchers should report evidence that instruments measure what they claim and do so consistently. Cognitive interviews support content validity. Test-retest reliability will be checked by asking 30 riders to complete the survey twice, one week apart, and comparing answers on helmet use and perceived risk. Self-reported helmet use may overstate actual use, so the observation tally will record helmet use among passing riders as a check.

Expected Limitations

Intercept surveys reach riders who park in busy areas and are willing to stop, which may underrepresent late-night riders and those in a hurry, groups that may be least likely to wear helmets. Answers about drinking may be underreported despite anonymity. These limits will be stated alongside results, and evening and late-night shifts are built into the sampling schedule to reduce the first.

Analysis

Results will be reported as percentages with confidence intervals, overall and by student status and riding frequency. Reasons for not wearing a helmet will be ranked. Willingness to use a supplied helmet will be compared between riders who never wear one and others.

A short summary will go to the city council and the scooter companies.

References

Creswell, J. W., & Creswell, J. D. (2018). Research design: Qualitative, quantitative, and mixed methods approaches (5th ed.). SAGE.

Dillman, D. A., Smyth, J. D., & Christian, L. M. (2014). Internet, phone, mail, and mixed-mode surveys: The tailored design method (4th ed.). Wiley.

Trivedi, T. K., Liu, C., Antonio, A. L. M., Wheaton, N., Kreger, V., Yap, A., Schriger, D., & Elmore, J. G. (2019). Injuries associated with standing electric scooter use. JAMA Network Open, 2(1), Article e187381. https://doi.org/10.1001/jamanetworkopen.2018.7381