Mixed methods: make numbers and accounts answer each other

Predict

Before calculating or reading feedback, predict what the evidence can establish and name one competing explanation.

A joint display aligns evidence around a shared question. Agreement, complementarity and disagreement are all informative. Interview accounts can suggest mechanisms but do not automatically estimate their population prevalence.

Inspect the authentic source

The NIH-commissioned guidance defines mixed methods through intentional integration of quantitative and qualitative evidence. Collecting both kinds without relating them does not by itself deliver an integrated explanation.

Source: Creswell, Klassen, Plano Clark and Smith; NIH OBSSR. Date: 2011 methodological report. This source note is an authored paraphrase.

Worked analysis

Synthetic access study: site A has 40 arrivals among 50 invitees; site B has 20 among 50. Four voluntary interviews at A mention convenient hours; four at B include three reports of shift conflicts and one of transport cost. Arrival rates are 80% and 40%. The interviews suggest explanations for investigation, but eight self-selected accounts cannot establish that shift conflict explains the 40-point gap.

Build and challenge your account

Build rows for hours, transport and other explanations, with columns for counts, coded accounts, fit or conflict, and next evidence. Draft an explanatory follow-up sample that includes attendees and nonattendees at both sites. Specify a consent and anonymization procedure without collecting classmates’ personal histories.

Use a separate line for each conclusion. Beside it, identify the source record, transformation or interpretive step, and assumption that could fail. Keep observations separate from explanations. If two interpretations fit the same record, identify an additional observation that would distinguish them rather than choosing the more familiar story.

Independent transfer

A survey reports high satisfaction but interviews describe access failures. What should integration do?

Write an answer before opening feedback. State what changes in the new setting and which part of the original reasoning still applies. A good transfer answer includes both a useful inference and a boundary on that inference.

Limits

The numeric and interview records are deliberately synthetic. The NIH report supplies methodological guidance, not evidence that this invented program exists or works. Small purposeful qualitative samples support depth and explanation, not automatic statistical generalization.

Review your evidence

Check the units and denominator of every number; check the author, date and access of every account. Distinguish an absent observation from evidence that an event did not occur. Retain a short revision note so a reviewer can follow how your conclusion changed.

Extend

High school: Produce the specified table, timeline or diagram; cite the source record and mark assumptions. Explain one tempting but unsupported conclusion.

AP-connected practice: Use this as interdisciplinary evidence practice where relevant to an existing AP course. Write a claim with specific evidence and a counter-explanation. This family is not represented as a separate AP course.

College: Extend the mechanism with a second authentic record selected for its ability to challenge your interpretation. Document sampling, measurement or source-access limits and a reproducible analysis trail.