Methods (Edexcel A-Level Psychology): Revision Notes
Observation and Research
Introduction to research methods in child psychology
Research in child psychology employs various methods to investigate how children develop and behave. Observational techniques, self-report methods such as questionnaires and interviews, and cross-cultural approaches each offer distinct ways to gather data about children's psychological development. These methods must be carefully selected and applied, particularly when working with vulnerable populations such as children, where ethical considerations are paramount.
Naturalistic and structured observations
Observational research involves watching and recording participant behaviour without directly manipulating variables. This approach falls into two main categories: naturalistic and structured observations.
Naturalistic observations
Naturalistic observation takes place in the participant's own natural environment. The researcher does not create or control the situation, allowing them to observe genuine behaviour as it occurs spontaneously. This method provides authentic insights into how people behave in their everyday contexts.
The primary strength of naturalistic observation is its high ecological validity. Because participants are in familiar surroundings, their behaviour is more likely to reflect how they typically act. For instance, Ainsworth conducted naturalistic observations in Uganda to investigate her Strange Situation hypothesis. Observing children in their home environments provided data that was representative of their normal behaviour patterns.
However, naturalistic observations present several challenges:
- Reliability issues: Since observations occur in natural, uncontrolled settings, it becomes difficult to replicate studies exactly. Each participant's environment differs, and spontaneous events cannot be recreated precisely.
- Control limitations: Researchers cannot control extraneous variables that might influence behaviour, making it harder to establish clear cause-and-effect relationships.
- Ethical concerns: When conducting naturalistic observations with children, obtaining proper informed consent is essential. Parents must fully understand what the research involves and why observations are taking place.
Ethical Considerations in Naturalistic Observations
When conducting naturalistic observations with children, obtaining proper informed consent is essential. Parents must fully understand what the research involves and why observations are taking place. Researchers have a responsibility to protect vulnerable participants throughout the observation process.
Structured observations
Structured observation involves staging situations within a controlled environment where the researcher maintains some degree of control. Behaviour can be observed through one-way mirrors or recorded on screens. These observations typically employ coding systems to record specific behaviours systematically.
Structured observations offer several advantages:
- Increased reliability: The use of standardised coding systems allows for replication. Different researchers can observe the same types of behaviours using identical criteria, making results more consistent.
- Access to specific behaviours: Researchers can create situations that elicit particular behaviours that might be difficult to observe naturally. Ainsworth used structured observations during the Strange Situation to examine specific attachment behaviours.
- Better control: The controlled environment reduces the influence of confounding variables, strengthening the validity of conclusions.
Nevertheless, structured observations have limitations:
- Reduced ecological validity: The artificial nature of the setting may cause participants to behave differently than they would in natural circumstances.
- Ethical considerations: Researchers must ensure children are not subjected to undue stress during structured observations. The procedures should not cause harm or distress.
The trade-off between naturalistic and structured observations often comes down to ecological validity versus control. Naturalistic observations provide authentic behaviour in natural settings, while structured observations offer better control and replicability. Researchers must carefully choose which approach best suits their research questions.
Quantitative and qualitative data from observations
Observations can generate both quantitative and qualitative data. Quantitative data involves numerical information, such as frequency counts of specific behaviours (tallying how many times a child displays a particular action). This type of data is objective and easily analysed statistically.
Qualitative data consists of descriptive information, such as written quotes or detailed descriptions of observed behaviours. Whilst this provides rich, detailed insights, it is more subjective and relies on the researcher's interpretation, making it harder to quantify numerically.
Questionnaires and interviews as research methods
Self-report methods, including questionnaires and interviews, centre on asking participants questions to gather information about their thoughts, feelings, or experiences.
Questionnaires in child psychology
Questionnaires consist of written questions that participants answer independently. They can be useful tools for assessing children's views, though they are more commonly used to gather information from parents and teachers about children's behaviour.
One major issue with questionnaires is social desirability bias – participants may respond in ways that present themselves more favourably rather than answering truthfully. This distorts the validity of the data collected. Respondents might exaggerate positive behaviours or underreport negative ones to create a better impression.
Limitations with Young Children
Questionnaires may not be suitable for young children who lack the reading comprehension or attention span required to complete them independently. Researchers must carefully consider the developmental capabilities of their participants when selecting research methods.
However, questionnaires can be effectively used with adolescents. The Child and Adolescent Survey of Experiences (CASE) provides an example of a questionnaire designed to measure stressful life experiences in this age group.
Interviews with children
Interviews are face-to-face interactions involving a series of questions. They allow researchers and participants to expand on topics and clarify responses, enabling more detailed data collection than questionnaires typically provide.
Special Considerations When Interviewing Children
When interviewing children, several important factors must be taken into account:
- Training requirements: Interviewers working with children need specific training to communicate effectively with this population.
- Language adaptation: The vocabulary and sentence structure must be adjusted to suit the child's developmental level. Language complexity changes according to the child's age.
- Attention span: Young children have limited attention spans, so expecting them to sit through lengthy interviews is inappropriate. Sessions should be brief and engaging.
- Recording methods: Many researchers record interviews and later transcribe them to identify common themes or denominators of research interest.
- Vulnerability considerations: Children represent a more vulnerable group, requiring interviewers to take extra care in their approach and questioning techniques.
Interviewer effect
The interviewer effect refers to how characteristics of the interviewer—including age, gender, and ethnicity—can influence participants' responses. This effect is particularly relevant in child psychology research.
The physical appearance and demeanour of an interviewer may shape how a child responds to questions, depending on the topic being discussed. Children might alter their answers to align with what they perceive the interviewer wants to hear. These subtle cues about expectations can be considered demand characteristics—hints that lead participants to modify their responses to conform to perceived expectations.
Impact on Research Validity
The interviewer effect profoundly affects the reliability and validity of results obtained. Researchers must be aware of how their own characteristics and behaviour might influence participant responses, particularly when working with children who may be more susceptible to these influences.
Cross-cultural research
Cross-cultural research enables psychologists to determine whether behaviours and psychological phenomena are universal across different countries and cultures, or whether they are culturally specific. Within child psychology, researchers can investigate whether behaviours such as attachment patterns are consistent regardless of cultural background.
If cross-cultural research is not conducted, findings would only be relevant to the specific culture studied, resulting in culturally biased conclusions. For instance, Takahashi (1990) sought to test whether the Strange Situation procedure, originally developed by Ainsworth and Bell (1970) in America, was valid for cultures beyond the United States.
Takahashi (1990) study
Worked Example: Cross-Cultural Validation Study
Takahashi observed 60 Japanese mothers and their children using the Strange Situation procedure to test its cross-cultural validity.
Key Findings:
- 68% of children were classified as securely attached, which was comparable to Ainsworth and Bell's findings
- No avoidant-insecure behaviour was observed
- 32% were classified as resistant
Cultural Interpretation:
Notably, Japanese children appeared very distressed when left alone, which the researcher attributed to cultural factors. Japanese culture emphasises that avoidant behaviour (avoiding interaction with others) is impolite and discouraged from an early age.
Conclusion:
This cultural teaching suggested that the Strange Situation does not measure attachment behaviour effectively on a universal scale, as cultural values influence how children respond to the separation and reunion episodes.
Cross-sectional and longitudinal designs
Researchers must decide whether to use cross-sectional or longitudinal designs when conducting studies, particularly in developmental psychology.
Cross-sectional designs gather information from a population at a single point in time. Researchers select a cross-section of the population to study and compare their measurements at one moment. For example, to examine how insecure attachments change over time, a researcher might assess a group of two-year-olds identified as insecurely attached and compare them with a group of four-year-olds with the same attachment style.
Advantages of Cross-Sectional Studies:
- Immediate results: Data can be collected quickly without waiting for participants to age.
- Cost-effective: Only one round of data collection is needed, reducing expenses.
- Fewer ethical concerns: Less burden is placed on participants compared to long-term studies.
However, cross-sectional studies have important limitations. Different participants may have experienced very different upbringings and life experiences, which could affect results even if they share the same attachment style. Additionally, researchers cannot track actual changes in the same individuals over time.
Longitudinal research follows the same participants over an extended period, collecting data at multiple time intervals. This approach allows researchers to observe whether behaviours remain consistent or change as individuals develop.
Advantages of Longitudinal Studies:
- Avoids cohort effect: The cohort effect refers to differences within social and cultural groups that change over time and with age. By following the same individuals, longitudinal studies eliminate this confound.
- Tracks individual change: Researchers can see exactly how each participant develops, providing more detailed insights into developmental trajectories.
Limitations of Longitudinal Studies:
- Time-consuming and expensive: Following participants over many years requires sustained funding and commitment.
- Difficult to replicate: The extended timeframe and specific conditions make replication extremely challenging.
- Attrition: Attrition is the tendency for participants to drop out of studies over time for various reasons (moving away, losing interest, life changes). This threatens the validity of the study and can alter the research direction or force researchers to find replacement samples.
Meta-analysis in cross-cultural research
Much cross-cultural research is conducted through meta-analysis. This technique involves combining and reanalysing results from multiple individual studies that investigated a specific topic through a statistical method. Meta-analysis allows researchers to develop a broader perspective by treating numerous smaller studies as a single, larger pool of data. This makes overall trends and patterns more likely to be identified with greater statistical power.
Van IJzendoorn and Kroonenberg (1988) conducted a meta-analysis examining attachment types across different countries. They analysed 32 studies that had used the Strange Situation to measure attachment in eight different countries. Their analysis found considerable consistency in the overall distribution of attachment types across cultures, suggesting some universality in attachment patterns whilst also revealing some cultural variations.
Key Points to Remember:
- Naturalistic observations occur in natural settings with high ecological validity but can be difficult to replicate, whilst structured observations provide better control and reliability but may lack ecological validity.
- Social desirability bias affects self-report methods when participants answer in ways that present themselves favourably rather than truthfully.
- The interviewer effect describes how interviewer characteristics (age, gender, ethnicity) can influence participant responses, particularly relevant when researching children.
- Cross-cultural research determines whether psychological phenomena are universal or culturally specific; failure to conduct such research results in culturally biased conclusions.
- Cross-sectional designs gather data at one time point and are cost-effective, whilst longitudinal designs follow participants over time but face challenges such as attrition and cost.