Human Research (Edexcel A-Level Psychology): Revision Notes
Human Research
Observational research methods in psychology
Observational research is a method where researchers collect data by watching and recording behaviour without manipulating any variables. This approach is widely used across psychology, from laboratory experiments to naturalistic case studies. Unlike experimental methods, there is no independent variable to manipulate—the researcher simply observes what naturally occurs.
Quantitative and qualitative data
When conducting observational research, psychologists can gather two main types of data:
Quantitative data refers to numerical information that can be counted or measured. For example, a researcher might record the number of aggressive acts displayed by children during playtime. This data can be statistically analysed and is objective in nature.
Qualitative data consists of descriptive information expressed through words, texts, ideas and themes. This type of data cannot be expressed numerically but provides rich, detailed insights into behaviour. For instance, a researcher might record detailed notes about the context and nature of children's interactions.
The choice between quantitative and qualitative data depends on the research question. Quantitative data allows for statistical analysis and objective comparisons, whilst qualitative data provides deeper understanding of the context and meaning behind behaviours.
Data collection techniques
Researchers employ several techniques to collect observational data systematically:
Tallying involves recording each occurrence of a specific behaviour by making a mark in the appropriate category. For example, Table 4.4 shows a tally chart used to observe whether males and females drive different sized cars. Each time an observer sees a person in a particular category (e.g., female driving a small car), they make a tally mark in that cell. This creates a simple frequency count of behaviours.
Time sampling requires observers to make recordings at specific time intervals, such as every 30 seconds. The observer records what behaviour is occurring at that precise moment. This method ensures systematic coverage of an observation period, though it may miss certain behaviours if they only occur between sampling points.
Event sampling involves recording a particular behaviour every time it occurs. For instance, a researcher might tick a box each time someone displays a specific behaviour, such as aggressive acts. The limitation here is that if the behaviour happens very frequently, the observer may struggle to record all instances accurately.
Worked Example: Choosing a Data Collection Technique
Scenario: A researcher wants to observe how often children share toys during a 30-minute play session.
Option 1: Tallying - Record every instance of toy-sharing behaviour throughout the entire session. Best if sharing occurs at a moderate frequency.
Option 2: Time sampling - Observe what the child is doing every 2 minutes (at 0:00, 0:02, 0:04, etc.). Record if sharing is occurring at that exact moment. Best for getting an overall picture of behaviour patterns.
Option 3: Event sampling - Record each complete sharing event from start to finish. Best if the researcher wants detailed information about the context of each sharing instance.
Standardisation and training
For observational research to be reliable, behavioural categories must be clearly defined and unambiguous. In large-scale studies, multiple observers require training to ensure they all understand the operational definitions of behavioural categories.
This standardisation is essential when more than one observer is making assessments, as it ensures consistency in how behaviours are identified and recorded. Without clear operational definitions, different observers may categorise the same behaviour differently, compromising the reliability of the research.
Types of observation
Observational methods can be classified according to several dimensions. Understanding these distinctions is essential for evaluating the appropriateness and validity of observational research.
Naturalistic vs structured observations
Naturalistic observation takes place in the participant's natural environment, where the situation has not been created or manipulated by the researcher. This approach provides genuine insight into how people behave in real-world contexts. For example, observing children's behaviour in a school playground during break time would be naturalistic. The authenticity of the setting enhances ecological validity.
Structured observations are staged observations conducted in a controlled environment, typically a laboratory setting. The researcher has control over certain aspects of the situation, and participants' behaviour can be observed behind a one-way screen. These observations are designed to record specific behaviours where it may be difficult to obtain information from naturalistic settings.
Structured observations generate numerical data, making them suitable for statistical analysis. They tend to be more reliable than naturalistic observations because the coding systems allow for replicability. However, the controlled setting may feel artificial, potentially affecting how naturally participants behave. Additionally, ensuring all observers interpret information consistently can be challenging, which is why having multiple observers helps validate the study.
The choice between naturalistic and structured observations involves a trade-off: naturalistic observations offer higher ecological validity but lower control and reliability, whilst structured observations offer higher control and reliability but potentially lower ecological validity.
Participant vs non-participant observations
Non-participant observation occurs when the researcher observes without becoming part of the group being studied. This maintains researcher objectivity and allows for detailed note-taking during the observation. However, the presence of an observer may influence participants' behaviour, potentially reducing validity. Non-participant observations also risk missing subtle data that might only be apparent from within the group.
Participant observation involves the researcher taking an active role in the situation being observed. By becoming part of the group, the researcher gains insider perspective and can access data that might be missed by external observation methods. This approach is particularly valuable for understanding group dynamics from within. However, recording observations while participating can be difficult, and the researcher's presence may still affect how participants behave. The obvious advantage is that there is less stranger anxiety since the observer is part of the group.
Overt vs covert observations
Overt observation occurs when those being observed are aware they are being watched. For instance, an inspector entering a classroom to observe teaching would be conducting overt observation. The ethical advantage is that informed consent can be obtained, and participants can be informed of their right to withdraw. However, awareness of being observed may cause participants to alter their behaviour, reducing the validity of the observations.
Covert observation takes place when participants are unaware they are being observed. An example would be observing student behaviour while sitting inconspicuously in a student area. Since participants are unaware of being watched, they are unlikely to change their behaviour, making the observations more valid and reflecting natural behaviour patterns. However, covert observation raises ethical concerns as participants cannot give informed consent and are unaware of their involvement in research.
Covert observation presents a significant ethical dilemma: whilst it produces more valid data by eliminating observer effects, it violates the principle of informed consent and may infringe on participants' privacy rights. Researchers must carefully weigh these considerations before choosing this method.
Evaluation considerations
When evaluating different observation types, several factors must be considered:
Ethical issues vary considerably. Overt observations allow for informed consent and respect participant autonomy, whilst covert observations may violate privacy and prevent withdrawal. Participant observations can create difficulties with recording notes and maintaining objectivity.
Subjectivity affects all observational methods. The researcher's personal interpretation and potential bias can influence what is recorded and how it is interpreted, particularly in qualitative observations.
Reliability and validity differ across methods. Structured observations with clear coding systems tend to be more reliable and replicable. Naturalistic observations often have higher ecological validity but may be less reliable. Multiple observers can enhance reliability through inter-rater reliability checks.
Observer effects occur when the presence of a researcher influences participant behaviour. This is particularly problematic in overt and participant observations but can be minimised in covert and non-participant approaches.
Key Points About Observation Types:
- Naturalistic observations offer high ecological validity but lower control
- Structured observations provide better reliability and control but may feel artificial
- Participant observations give insider perspective but make recording difficult
- Non-participant observations allow objectivity but may miss subtle details
- Overt observations are ethical but may cause behaviour changes
- Covert observations produce valid data but raise ethical concerns
Content analysis
Content analysis is a research technique used to analyse the occurrence of specific words, images or concepts within material such as advertisements, books, films, newspapers or other media. This method allows researchers to examine patterns and trends in communication systematically.
Procedure
A researcher using content analysis begins by identifying the material to be analysed and establishing clear categories in advance. For instance, when investigating gender stereotyping in children's books, categories might include instances where males and females display stereotypical versus non-stereotypical behaviours. The researcher then examines the selected material systematically, tallying or counting occurrences within each category.
Content analysis can employ both quantitative and qualitative approaches. Quantitative analysis involves counting frequencies of particular categories. Qualitative analysis examines the meanings and relationships between words, concepts and images, making inferences about the messages conveyed. It is essential that content is coded into clear and manageable categories to enable appropriate conclusions.
Worked Example: Content Analysis of News Coverage
Research question: Do newspapers show gender bias in reporting sports achievements?
Step 1: Select material - Choose 50 recent newspaper articles about sports achievements (25 about male athletes, 25 about female athletes)
Step 2: Establish categories:
- Language used: aggressive/competitive vs graceful/elegant descriptions
- Focus on appearance vs performance
- Mention of family/personal life vs professional achievements
Step 3: Code the content - Read each article and tally instances in each category
Step 4: Analyse patterns - Compare frequencies between articles about male and female athletes to identify potential bias
Step 5: Draw conclusions based on quantitative frequencies and qualitative themes identified
Evaluation
Strengths:
Content analysis is an unobtrusive method that rarely raises ethical concerns since data is collected from existing sources rather than live participants. Researchers should nevertheless respect confidentiality when analysing materials. The method offers opportunities for fresh interpretation of existing data, which may not be achievable through other approaches. It proves particularly valuable for analysing historical material and documenting trends over extended periods. Reliability can be easily assessed because content analysis is straightforward to replicate using the same categories and materials.
Weaknesses:
The method relies heavily on the researcher's personal interpretation, introducing an element of subjectivity or bias. Internal validity may be compromised if the categories chosen do not accurately represent what they claim to measure, rendering the data invalid. Content analysis is essentially a descriptive method—it reveals what patterns exist but cannot explain the underlying motives behind observed patterns. The method is limited to available material, and observed media trends may not accurately reflect reality. For example, dramatic events typically receive more media coverage than routine occurrences, potentially distorting the picture of actual frequency or importance.
Remember!
Key Concepts to Remember:
-
Observational research involves gathering data by watching behaviour without manipulating variables. It can collect both quantitative (numerical) and qualitative (descriptive) data through techniques like tallying, time sampling and event sampling.
-
Six main observation types exist: naturalistic (natural environment) vs structured (controlled setting); participant (researcher joins group) vs non-participant (researcher remains separate); and overt (participants aware) vs covert (participants unaware). Each has distinct ethical and validity implications.
-
Content analysis is a research technique for systematically analysing words, images or concepts in media materials. It can be highly reliable and useful for historical analysis but relies on researcher interpretation and cannot explain underlying causes.
-
Evaluation considerations for all methods include ethical issues, reliability and validity, subjectivity, observer effects, and the distinction between describing patterns (what happens) versus explaining causes (why it happens).