Research methods in developmental psychology: clinical interviews, ethnography, longitudinal and cross-sectional designs

Clinical interviewing

A clinical interview is a flexible, semi-structured conversation in which the researcher starts with set questions or a task, then follows up the child's answers with further questions to understand how the child is thinking. Piaget used it: he set a conservation task, then asked 'Why do you think that?' and adapted his next question to the answer.

• Strengths: rich qualitative data about the child's own reasoning; flexible, so it can follow surprising answers; can be adapted to the child's language and age.
• Weaknesses: each interview differs, so it is hard to replicate and compare (lower reliability); the interviewer's questions may lead the child (children may change answers when a question is repeated, as in conservation studies); interpretation is subjective; young children may lack the language to explain.

Ethnographic fieldwork: Punch (2002)

Ethnography means studying a group in its natural setting over a long period, often living alongside them, using participant observation, informal interviews and other methods to understand their world from the inside.

Punch (2002), 'Research with children: the same or different from research with adults?' (*Childhood*), drew on fieldwork in a remote rural community in Bolivia with no electricity. She worked with 37 children aged 8 to 14, using semi-participant observation and informal interviews, plus task-based methods: drawings, photographs (17 children took photos), diaries (over half kept them for more than two months), spider diagrams and activity tables, and worksheets. Parents and teachers were also interviewed.

She argued research with children differs from research with adults mainly because of adults' perceptions of children and children's marginalised position, not because children are inherently different. Power imbalances make trust slow to build. 'Child-friendly' methods are not automatically better, so a mix of traditional and innovative methods helps engagement, data quality and triangulation; she suggested 'research participant-centred' methods instead of 'child-centred'.

Longitudinal and cross-sectional designs

Evaluating developmental methods

Use the terms the specification lists: reliability (could it be repeated with the same results?), validity (does it measure real development?), generalisability, credibility, objectivity and subjectivity, ethics (children are vulnerable participants) and practical application. A good answer links the method to the specific study or scenario.

Worked example

A researcher wants to know whether children's friendships become more stable between ages 5 and 10, and has one year to finish. Which design should she use, and what is the trade-off?

1. One year is too short to follow children from 5 to 10, so a **cross-sectional** design (compare groups of 5, 7 and 10-year-olds now) is practical.
2. Trade-off: differences may be **cohort effects**, and she cannot see how individual children change.
3. A longitudinal design would show real change but would take five years and risk attrition.

Answer: Cross-sectional, accepting cohort effects instead of tracking individuals.

Check yourself

1. Piaget asks a child a conservation question and then asks follow-up questions depending on her answer. This is
   • a clinical interview
   • a structured questionnaire
   • a content analysis
   • a cross-sectional design
   Show answer

   a clinical interview. Flexible questioning that follows the child's reasoning.

2. What is ethnographic fieldwork?
   • Testing children in a laboratory for one hour
   • Pooling data from many studies
   • Studying a group in its natural setting over a long period to understand their world from the inside
   • Comparing age groups at one point in time
   Show answer

   Studying a group in its natural setting over a long period to understand their world from the inside. Long-term immersion in the setting.

3. Punch (2002) worked with 37 children, of whom 17 took photographs. What percentage took photographs? Give your answer to 1 decimal place.
   Show answer

   45.9 %. 17 / 37 x 100 = 45.9%.

4. What did Punch (2002) conclude about research with children?
   • Children must only be studied with drawings
   • Research with children is identical to research with adults
   • Children cannot take part in research
   • Differences from adult research come mainly from adults' perceptions and children's position in society, and a mix of methods works best
   Show answer

   Differences from adult research come mainly from adults' perceptions and children's position in society, and a mix of methods works best. She argued for a continuum and a mix of methods.

5. A cross-sectional study finds 70-year-olds are worse with computers than 20-year-olds. The most likely problem is
   • attrition
   • a cohort effect
   • practice effects
   • order effects
   Show answer

   a cohort effect. The groups grew up with different technology.

6. Which weakness is most typical of longitudinal designs?
   • Cohort effects between age groups
   • No data on change
   • They are always quicker than cross-sectional studies
   • Participants drop out over time, which may bias the sample
   Show answer

   Participants drop out over time, which may bias the sample. Attrition.

7. Explain one weakness of using clinical interviews with children.
   Show answer

   Because questions change for each child, interviews are hard to replicate and compare, so reliability is low; the interviewer may also lead the child with questions.. Name the weakness and why it happens.

Practise with instant feedback Print this lesson

Sources

• Pearson Edexcel International Advanced Subsidiary/Advanced Level in Psychology specification, Issue 6 (November 2024, first assessment January 2026)
• Punch (2002), Research with children: the same or different from research with adults?, Childhood 9(3)
• OpenStax Psychology: 9.2 Lifespan theories (Piaget, Erikson)
• Pearson IAL Psychology summary of studies: WPS03

Last updated 8 November 2026. Spotted a mistake? Tell us.