Analyzing Data
- Distinguish descriptive from inferential statistics and interpret basic measures
- Outline the process of coding and thematic analysis for qualitative data
- Explain why correlation does not establish causation
Quantitative analysis: describe, then infer
Quantitative analysis has two levels. Descriptive statistics summarize your data: measures of central tendency (mean, median, mode) and of spread (range, standard deviation), plus frequencies and percentages. Inferential statistics use a sample to draw conclusions about a larger population and to judge whether a pattern is likely real or due to chance — reported with p-values and significance tests, or with correlation coefficients that measure how strongly two variables move together. You do not need advanced statistics for AP Research, but you must interpret your chosen measures correctly and not overstate what they show.
Qualitative analysis: coding and themes
Qualitative data — interview transcripts, open responses, field notes — is analyzed by coding: reading closely and tagging segments with labels that capture their meaning. Repeated codes are grouped into themes, and themes are interpreted in light of your research question. This is thematic analysis, and its rigor comes from being systematic and transparent: applying codes consistently, keeping an audit trail, and often having a second coder or using member checking and triangulation to guard against the researcher simply seeing what they expected. The goal is a defensible interpretation, not a cherry-picked quote.
Correlation is not causation
The most important interpretive rule in all of research: a correlation between two variables does not prove that one causes the other. Two variables can move together because A causes B, because B causes A (reverse causation), or because a hidden third variable (a confound) drives both — or by pure coincidence. Only a well-controlled experiment with a manipulated variable and a control group can support a causal claim. When your data is correlational (as most survey data is), write in the language of association — "is related to," "is associated with" — not causation.
A student’s survey finds that students who report more sleep also report higher grades, and writes, "This proves that sleeping more causes better grades." Evaluate and correct the interpretation.
- 1.Identify the data type: this is a correlational survey — sleep and grades were measured, not manipulated.
- 2.Recognize the overclaim: "proves" and "causes" assert causation from a correlation.
- 3.Generate alternatives: reverse causation (organized, high-achieving students may manage time to sleep more) and confounds (family stability, workload) could drive both.
- 4.Note the design limit: without an experiment controlling other factors, causation cannot be established.
- 5.Correct the language: "Students who reported more sleep tended to report higher grades; the two are positively associated, though this study cannot establish that one causes the other."
A study finds a strong positive correlation between hours of TV watched and a child’s shoe size. Which conclusion is most defensible?
Most student survey data is correlational, so causal language ("causes," "leads to," "proves") is almost always an overclaim. Default to association language — "is related to," "is associated with" — unless you ran a controlled experiment.
In qualitative analysis, what is the purpose of "coding" interview transcripts?
Report analysis at a level you can defend. Reviewers reward interpretation matched to your data and method — correct descriptive statistics and honest association language, or transparent coding and themes — far more than an impressive-sounding test you cannot justify or a causal claim your design cannot support.
Answer the 2 checkpoints as you read.
Sign in to save your progress