DDBA 8300 Week 4 Research Question Audit Example

Two Studies Hiding in One Sentence: Auditing a Research Question About Free Returns, Repeat Purchases and How Shoppers Feel About Sending Clothes Back, Through Four Rewrites Until One Design Fits

Student Name

Walden University

DDBA 8300: Qualitative and Quantitative Methodology for Applied Business Research

Instructor Name

Month Day, Year

Two Studies Hiding in One Sentence: Auditing a Research Question About Free Returns, Repeat Purchases and How Shoppers Feel About Sending Clothes Back, Through Four Rewrites Until One Design Fits

The Draft Question

My study concerns a composite online women's clothing retailer based in Columbus, Ohio, that switched from charging $7.95 for returns to offering free returns eighteen months ago. Returns rose, and the owner wants to know whether free returns are paying for themselves through customer loyalty. My first draft of the research question read:

"What is the impact of free returns on customer loyalty, and how do shoppers feel about returning clothes they bought online?"

The sentence reads well. It also contains two studies, and neither is designed well by its wording.

Rewrite One: Separating the Two Questions

"What is the impact of free returns on customer loyalty?" and, separately, "How do shoppers feel about returning clothes they bought online?"

Removed: the conjunction joining two questions that need different designs. The first asks about a relationship between a policy and a behavior, which calls for measurement across many customers. The second asks about experience, which calls for interviews. A single study answering both would need to be mixed methods, which the course allows, but the draft gave no reason for mixing and no plan for how the parts would connect. Alvesson and Sandberg (2011) note that research questions often bundle assumptions without examining them; here, the bundled assumption was that feelings about returning explain loyalty, which neither half of the question tests. I chose to pursue the first question and set the second aside for a possible later study.

Rewrite Two: Removing "Impact"

"What is the relationship between free returns and customer loyalty?"

Removed: "impact," which implies a causal effect. The retailer changed its policy for all customers at once, so there is no group that kept paid returns. A design comparing customers before and after the change could show an association but could not separate the policy's effect from everything else that changed over eighteen months, including a new website and higher advertising spending. Creswell and Creswell (2018) distinguish questions about relationships from questions about effects and match each to designs that can support it. "Relationship" is the honest word.

Rewrite Three: Defining "Loyalty"

"What is the relationship between customers' use of free returns and their repeat purchases within twelve months of their first order?"

Removed: "loyalty," which could mean attitude, intention or behavior. The retailer's records hold behavior: orders, returns and dates for every customer. Repeat purchase within twelve months is measurable for all customers who placed a first order in the period. "Free returns" also needed definition; the policy applies to everyone, so the variable that varies is whether and how much a customer used it. Petersen and Kumar (2009), studying a catalog retailer's records, found that customers who returned some products went on to buy more in later periods, up to a point. That finding suggests the relationship is worth testing and may not be simple.

Rewrite Four: Bounding the Population and Adding Controls

"Among customers who placed a first order between January and June of last year, what is the relationship between the number of items returned in the first ninety days and the number of orders placed in the following twelve months, accounting for first-order size and product category?"

Removed: the unbounded population. The question now names who is studied, when and over what windows, and names the controls the analysis needs, since customers who order more items also return more.

Checking the Final Question Against the Records

Before settling on the final wording, I asked the retailer's analyst to confirm that every term could be pulled from the order system. First-order dates, item counts, returns within ninety days, product categories and later orders are all stored with customer identifiers, and about 14,000 customers placed a first order in the window. Two terms needed adjustment: returns initiated but never received are recorded separately, so the question counts only returns received, and exchanges are counted as returns, since the policy covers both. A question whose terms cannot be pulled from the data is not finished, however clear it reads.

Which Design Fits

The final question fits a correlational design using the retailer's transaction records for about 14,000 first-time customers. Edmondson and McManus (2007) describe methodological fit as consistency between the question, prior knowledge and design; the question now asks about a relationship between measured variables that prior research has already studied in another setting, which is the condition under which a quantitative design fits.

What Was Lost

The final question no longer asks how shoppers feel. That was the part I found most interesting, and it may be what the owner most wants to understand. But it belongs to a different study. Keeping it in the same sentence would have produced a design that answered neither question well.

References

Alvesson, M., & Sandberg, J. (2011). Generating research questions through problematization. Academy of Management Review, 36(2), 247-271. https://doi.org/10.5465/amr.2009.0188

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

Edmondson, A. C., & McManus, S. E. (2007). Methodological fit in management field research. Academy of Management Review, 32(4), 1246-1264. https://doi.org/10.5465/amr.2007.26586086

Petersen, J. A., & Kumar, V. (2009). Are product returns a necessary evil? Antecedents and consequences. Journal of Marketing, 73(3), 35-51. https://doi.org/10.1509/jmkg.73.3.035