Chapter 24 — Applied research methods
Lesson map
Lesson 24 published. Map, example, EX070.
flowchart TD
N0["Learning outcomes"]
N1["24.1 Framing a question"]
N2["24.2 Ethical data collection"]
N3["24.3 Cautious conclusions"]
N4["Synthesis"]
N0 --> N1 --> N2 --> N3 --> N4
N4 --> EX["Worked example"]
EX --> Q["Exercise EX070"]
Logical drawing
flowchart LR
N0["Learning outcomes"]
N1["24.1 Framing a question"]
N2["24.2 Ethical data collection"]
N3["24.3 Cautious conclusions"]
N4["Synthesis"]
N0 --> N1 --> N2 --> N3 --> N4 --> Q["EX070"]
[Learning outcomes]
|
v
[24.1 Framing a question]
|
v
[24.2 Ethical data collection]
|
v
[24.3 Cautious conclusions]
|
v
[Synthesis]
|
v
[Worked example] --> [Exercise EX070]
Learning outcomes
You will (1) frame a small applied research question; (2) collect data ethically from systems you are authorised to use; (3) draw cautious conclusions.
24.1 Framing a question
Rosa's programme requires a small applied study. She asks: how long do council certificates take to renew after alerts, and which factors predict delays? The question is answerable with data she can access, useful to the council and narrow enough for one term. Her tutor, Dr Imogen Hale, helps her rewrite vague aims such as "study PKI" into a question with measurable variables: alert-to-renewal time, certificate type, owner team and whether renewal is automated. She also writes down what result would surprise her, which helps her notice if she is only looking for confirmation. Dr Hale asks her to list the data she cannot obtain, so the limits are known from the start.
24.2 Ethical data collection
Rosa obtains written permission from Owen to use ticket and monitoring data. She removes staff names, replacing them with team labels, and stores data on council systems only. She collects no citizen data, since it is not needed. Her tutor checks the plan against the polytechnic's research ethics guidance. Rosa learns that even operational data about colleagues deserves care, because results could embarrass individuals if published carelessly. She agrees with Owen that the council may review the report before submission, to catch anything that could identify individuals or reveal sensitive configuration. Raw data is deleted at the end of term, as the plan promised.
24.3 Cautious conclusions
Across 41 renewals, automated ones completed within a day after alerts, while manual ones took a median of six days. Rosa concludes that automation is associated with faster renewal, but cautions that teams with automation may also be better resourced, so the study shows association, not proof of cause. She reports how many cases were excluded for missing data and why. Her tutor praises the restraint and suggests a follow-up comparing teams before and after automation. She also publishes her analysis script with synthetic sample data, so others can repeat the method.
Worked example
Rosa studies 41 renewals, finds automated renewals finish within a day versus a six-day manual median, and reports association rather than cause.
Synthetic case. The study and figures are invented.
Errors and limits
Small samples give unstable results. Observational data cannot prove cause without careful design.
Exercises
- EX070 — Research question. What was Rosa's research question? Rubric: 3 = how long renewals take after alerts and which factors predict delay;; 2 = question;; 1 = topic only;; 0 = none.
- EX071 — Ethics step. Name two ethical steps Rosa took. Rubric: 3 = two of written permission, staff names replaced, data kept on council systems, no citizen data, ethics check;; 2 = one;; 1 = vague;; 0 = none.
- EX072 — Conclusion limit. Why can't Rosa claim automation causes faster renewal? Rubric: 3 = automated teams may also be better resourced; association only;; 2 = confounder;; 1 = can claim;; 0 = none.
Synthesis
Applied research needs a narrow question, ethical data handling and conclusions limited by design.
Continue studying — chapter 24 of 30
← Chapter 23 (draft) · Course chapter index · Chapter 25 (draft) →