HD560 Unit 6 Seminar Reflection Example

The Drop-In Hour I Did Not Count: A Unit 6 Seminar Reflection on a Rival Explanation Inside a Randomized Design

[Student Name]

Purdue University Global

HD560: Health Education Evaluation and Research

Unit 6 Seminar Reflection

[Instructor Name]

[Date]

The seminar, the classmates and the Within Earshot evaluation are composite; the validity framework and trial cited are real.

What I Brought

I presented my design with some pride. Applicants are assigned by lottery, both groups are assessed by the same people on the same schedule, and the primary outcome is a well-tested loneliness scale. When our instructor asked each of us to name the strongest rival explanation in our own design, I said I could not find one. Randomization, I said, had handled selection, history and regression toward the mean.

The Question

Rosa asked a practical question: after the aids are fitted, what happens? I described the three adjustment visits at the senior center. She asked whether there was anything else. I mentioned the monthly drop-in hour, where participants bring their aids for cleaning and battery help, and where, I added, many stay for coffee.

Then she asked the question that mattered. If loneliness falls in the group offered the program, how would I know whether it fell because people hear better or because, for six months, they had at least four reasons to leave the house and sit with other people?

I had no answer. The studies linking hearing loss to loneliness mostly describe associations (Shukla et al., 2020), and they cannot tell me which part of a program would carry an effect either. The lottery guarantees that the difference between groups is caused by the offer. It says nothing about which part of the offer caused it.

Randomization had protected my comparison, and I had let that protection stand in for an understanding of what was being compared, when the program in the treatment group was hearing aids plus regular company.

Why I Missed It

Looking back, I think I treated the program as its name. "Free hearing aids" sounded like a device, so I thought only about hearing. The drop-in hour was in the program manual, but I had filed it under logistics. Shadish et al. (2002) describe this as a construct validity problem: the label given to a treatment may not match all of what was delivered. I had studied that idea and still did not see it in my own design, because the label was mine.

I also wanted the design to be finished. A randomized design feels complete, and that feeling discouraged me from asking further questions.

What I Considered

The class discussed three responses. The first was an attention-control group: give wait-listed applicants a monthly social hour of equal length. A large hearing trial did something similar, setting its hearing intervention against a successful-aging education program that gave participants comparable time with staff (Lin et al., 2023). I decided against it, because the agency cannot staff a second program, and because the board's decision is about the program as delivered, social contact included.

The second was to remove the drop-in hour. That would make the evaluation cleaner but the program worse, and I rejected it.

The third was to measure the pathways. I chose this one.

What I Changed

The plan now records, for each person offered the program, the visits attended and drop-in hours. The analysis will check how much of any fall in loneliness tracks the fall in hearing handicap, and how much tracks drop-in attendance. I have relabeled the primary result as the effect of being offered the program as delivered, including its social contact. Interviews with participants will ask directly what they valued about the visits.

How It Felt

I was embarrassed for about ten minutes, mostly because I had announced so confidently that I could find no rival explanation. Then I was relieved that a classmate found it now rather than a reviewer after the results came in.

One Point I Still Hold

A classmate argued that the evaluation would be meaningless without an attention control. I disagree. The board is not deciding whether hearing aids reduce loneliness in general; it is deciding whether to keep this program. The comparison it needs is the one the lottery provides, honestly labeled.

What the Exchange Taught Me About Seminars

Rosa's question was not technical. She did not ask about statistics or sampling. She asked what happens to participants on an ordinary week, and that plain question found the problem my methodological checks had missed. I want to ask that question of my own designs, and of others', more often.

What I Take Forward

Before calling any design free of rival explanations, I now list everything participants in each group actually experience, not just the component in the program's name.

References

Lin, F. R., Pike, J. R., Albert, M. S., Arnold, M., Burgard, S., Chisolm, T., Couper, D., Deal, J. A., Goman, A. M., Glynn, N. W., Gmelin, T., Gravens-Mueller, L., Hayden, K. M., Huang, A. R., Knopman, D., Mitchell, C. M., Mosley, T., Pankow, J. S., Reed, N. S., ... Coresh, J. (2023). Hearing intervention versus health education control to reduce cognitive decline in older adults with hearing loss in the USA (ACHIEVE): A multicentre, randomised controlled trial. The Lancet, 402(10404), 786-797. https://doi.org/10.1016/S0140-6736(23)01406-X

Shadish, W. R., Cook, T. D., & Campbell, D. T. (2002). Experimental and quasi-experimental designs for generalized causal inference. Houghton Mifflin.

Shukla, A., Harper, M., Pedersen, E., Goman, A., Suen, J. J., Price, C., Applebaum, J., Hoyer, M., Lin, F. R., & Reed, N. S. (2020). Hearing loss, loneliness, and social isolation: A systematic review. Otolaryngology-Head and Neck Surgery, 162(5), 622-633. https://doi.org/10.1177/0194599820910377