Part 3: Reviewing a Student Housing Survey

Tutor: Right, let's look at this housing survey you've designed. I've gone through the draft, and there are a few areas we need to address before you start collecting data. What are your initial thoughts? Student A: Well, we tried to cover all the key aspects-rent, location, facilities, and satisfaction. But I'm a bit worried about the sample size. We were planning to survey 50 students, but that might not be enough to get reliable results. Student B: Actually, I thought 50 was fine. Most of the studies I've read use around 30 to 40 participants. But maybe we should check the expected response rate. If only half respond, we'll end up with 25, which is too low. Tutor: That's a valid point. The sample size needs to be large enough to allow for non-responses. I'd recommend aiming for at least 80 distributed questionnaires, so even with a 50% return, you'd have 40. But more importantly, how did you select your sample? Student A: We were going to use convenience sampling-just approach students in the library and cafeteria. But I've read that can introduce bias. Tutor: Exactly. You might end up with a sample that isn't representative. For instance, students who use the library might have different housing preferences than those who don't. Have you considered stratified sampling? Student B: We discussed it, but it seemed complicated. We'd need to divide the population into groups, like by year of study or type of accommodation, and then sample from each. But it would give more balanced results. Tutor: It's worth the effort. Now, let's look at the questions themselves. Question 5 asks: 'How satisfied are you with your accommodation?' and gives options from 'very satisfied' to 'very dissatisfied'. What's the issue there? Student A: It's a bit vague. Satisfaction could mean different things-like the physical condition, the landlord, or the value for money. We should be more specific. Student B: Maybe we could split it into several questions, each focusing on a different aspect. But then the survey gets longer, and people might not finish it. Tutor: That's a trade-off. Another problem is with Question 8: 'Do you think the rent is too high?' That's a leading question-it assumes the rent is high. How could you rephrase it? Student A: We could ask: 'How would you rate the affordability of your rent?' with a scale from 'very affordable' to 'very unaffordable'. That's more neutral. Tutor: Good. Now, about the interview plan. You mentioned doing some follow-up interviews. How will you select participants for those? Student B: We thought we'd ask at the end of the questionnaire if they're willing to be interviewed, and then contact a random sample of those who agree. Tutor: That's acceptable, but you need to ensure confidentiality. In the ethics section of your proposal, you must explain how you'll anonymise the data. Also, you'll need consent forms for the interviews. Student A: Right, we've drafted a consent form, but we weren't sure about how long to keep the recordings. Is there a standard period? Tutor: Usually, you'd keep them until the project is completed and then destroy them. Check the university guidelines. Now, for the final report, you'll need to present your findings clearly. What kind of graph do you plan to use for the satisfaction ratings? Student B: I thought a pie chart would be good-it shows proportions well. But my partner thinks a bar chart is better because we have multiple categories to compare. Tutor: A bar chart is generally more effective for comparing across categories, especially if you want to show differences between groups, like first-years versus final-years. You could also use a stacked bar chart to show the distribution of responses within each group. Student A: That's a good idea. And for the report structure, we were thinking of starting with an introduction, then methodology, results, discussion, and conclusion. Tutor: That's standard. But make sure the discussion section doesn't just repeat the results-it should interpret them and link back to the literature. Also, include a section on limitations. Your small sample size and sampling method will be key limitations to acknowledge. Student B: So we should be honest about the weaknesses. That makes sense. I think we have a lot to revise, but it's clearer now. Tutor: Good. Why don't you divide the tasks? One of you can work on the sampling and ethics, and the other on the questionnaire design and data analysis plan. Then we'll meet again next week. Student A: I'll take the sampling and ethics. I've been reading about stratified sampling, so I can work on that. Student B: And I'll revise the questions and plan the graphs. I'll also draft the interview questions. Tutor: Excellent. Remember, the key is to make the survey as objective and representative as possible. Let's aim for a pilot test with 10 students first to catch any remaining issues.