Guide to the Practical Investigation (Edexcel A-Level Psychology): Revision Notes

Guide to the Practical Investigation

Introduction

The practical investigation in social psychology involves conducting a questionnaire-based study rather than an experiment. Many classic studies in social psychology cannot be ethically replicated due to their potential to cause anxiety or embarrassment. Instead, you will investigate people's opinions and beliefs about behaviour by gathering both numerical and descriptive data through carefully designed questionnaires.

Learning outcomes

By completing the practical investigation, you will develop skills in:

• Designing questionnaires that collect both quantitative (numerical) and qualitative (descriptive) data on a social psychology topic
• Planning your questionnaire methodology, including appropriate sampling techniques and ethical safeguards
• Presenting collected data using tables, graphs, and descriptive statistics
• Conducting thematic analysis on qualitative responses to identify patterns and draw conclusions
• Evaluating the methodological strengths and limitations of your questionnaire approach
• Following standard reporting conventions when documenting your procedure, findings, and discussion

Developing your research aim

An aim identifies the general area of interest your questionnaire addresses. In the example practical investigation, the aim focuses on investigating perceptions of gender differences in obedience. Well-formulated aims derive from previous research in the area.

Before finalizing your study aim, conduct a thorough review of background theory and existing research. This ensures your investigation builds on established findings and addresses meaningful questions. For instance, obedience research suggests no actual differences in obedience levels between males and females, but females report greater distress when complying with orders they would not voluntarily follow. This background shapes an aim exploring perceived gender differences in obedience.

Formulating alternative hypotheses

Your practical investigation requires an alternative hypothesis – a prediction about your expected findings based on previous studies. This hypothesis may be directional or non-directional, depending on whether existing research provides clear evidence about the direction of expected differences.

Directional hypotheses

A directional hypothesis predicts the specific direction of the expected difference. Use this format when prior research consistently points to a particular outcome.

Non-directional hypotheses

A non-directional hypothesis predicts a difference without specifying its direction. Choose this format when previous research yields mixed or unclear findings, making it impossible to confidently predict which direction the difference will take.

Collecting quantitative and qualitative data

Your practical investigation should gather both quantitative data (numerical measurements) and qualitative data (detailed descriptive information). This mixed-methods approach provides both statistical evidence and rich insights into participants' attitudes.

Quantitative data through closed questions

Closed questions generate quantitative data by offering participants predetermined response options. These questions are straightforward to analyze because each response receives a numerical score.

Consider these examples of closed questions measuring perceived obedience:

1. "If a policeman asked a passer-by to pick up litter, who do you think would be more likely to comply with the request? man/woman"

2. "How likely would a man be to agree to make a colleague at work redundant if asked by his boss? very unlikely / unlikely / neutral / likely / very likely"

3. "A female student is asked by her teacher to photocopy a pile of work, but she does not have the time. How likely is the student to say no to her teacher? very unlikely / unlikely / neutral / likely / very likely"

Each response option receives a numerical value. In the second question, responses might be scored 1–5, where 1 represents "very unlikely" and 5 represents "very likely". Higher scores indicate greater perceived obedience. Totaling scores across multiple questions produces an overall obedience perception rating for each participant.

Qualitative data through open-ended questions and story completion

Qualitative data provides deeper insight into underlying attitudes that participants might not explicitly state when answering direct questions. Two main approaches gather this type of data:

Story completion tasks ask participants to finish a scenario, revealing their implicit beliefs about behaviour. In the example investigation, participants receive this incomplete scenario:

Open-ended questions directly ask for participants' views without constraining their responses to predetermined categories. Examples include:

1. "In your view, explain whether you think males or females would be more obedient in their workplace."

2. "Soldiers often have to comply with their superior officers' demands, frequently doing things they would not voluntarily do. Based on this, explain whether you think soldiers should be male or female."

These questions generate rich descriptive data about participants' reasoning and beliefs. However, they require more time to complete and analyze than closed questions.

Making sampling decisions

Before selecting your sampling method, identify the target population you wish to investigate. This may be the general population or a specific group, such as families, friends, or students.

For the example practical investigating perceptions of gender and obedience, the target population should include equal numbers of males and females. This balance is essential because perceptions of gender differences vary according to participants' own gender, age, and background. Ideally, use a random sample from a diverse target population to control for these individual differences, though this approach can be difficult and time-consuming.

Stratified sampling technique

A more practical approach employs stratified sampling, which ensures proportionate representation of key demographic characteristics. This technique maintains the gender balance necessary for the investigation while remaining feasible to implement.

Addressing ethical issues

Questionnaire design and distribution require careful attention to ethical principles. When planning your practical investigation, address these key considerations:

These ethical issues must be addressed before distributing your questionnaire. Poor question phrasing can inadvertently cause embarrassment or distress. Conduct a pilot study with a small sample of family and friends to identify any ethical concerns and ensure questions are clear and understandable. This preliminary testing allows you to refine problematic elements before wider distribution.

Sample participant brief

A well-designed participant brief might state:

"Thank you for agreeing to participate in a questionnaire for my psychology investigation. I am interested in how you view obedience in certain situations, so the questions you will answer are based on incidents involving someone giving orders. For most questions, you will need to decide whether you think a person will obey these orders. You will not be asked any personal questions and you are free to refuse to answer any or all questions.

In the second half of the questionnaire, you will be given the beginning of a story that you will be asked to complete.

Please answer as honestly as possible and complete the questionnaire independently without influence from others.

It will take approximately 20 minutes to complete the questionnaire, which will be completely anonymous. You will not provide your name, only your gender, and you can return the questionnaire to me in the pre-paid postal envelope. Your results and a full write-up of the findings will be available on request. After one week, the data will be extracted from your questionnaire response and your questionnaire will be destroyed."

Analyzing quantitative data

Once you have collected completed questionnaires, begin analyzing your quantitative and qualitative data. Closed questions provide numerical data that can be presented in tables and graphs, then summarized using descriptive statistics.

Data presentation in tables

Organize raw data systematically in tables showing each participant's scores. The table below illustrates how data might be structured using scores from Likert-style questions:

Participant	Score for perception of male obedience (total)	Score for perception of female obedience (total)
1	25	35
2	20	34
3	14	27
4	22	29
5	19	33
6	17	24
7	21	27
8	21	30
9	18	32

Examining this table reveals a clear pattern: participants' perceptions consistently rate females as more obedient than males. To understand this trend more precisely, calculate descriptive statistics.

Descriptive statistics

Descriptive statistics summarize data sets numerically, making patterns easier to identify and interpret. Calculate these key measures:

Mean: The arithmetic average, calculated by summing all scores and dividing by the number of scores. In this example:

• Mean perception of male obedience: $\bar{x} = 19.7$
• Mean perception of female obedience: $\bar{x} = 30.1$

The substantial difference between means (10.4 points) suggests participants perceive females as more obedient than males.

Mode: The most frequently occurring score in the data set.

• Mode for male obedience: 21
• Mode for female obedience: 27

Median: The middle value when scores are arranged in order.

• Median for male obedience: 21
• Median for female obedience: 30

Range: The difference between the highest and lowest scores, indicating data spread.

• Range for male obedience: 11
• Range for female obedience: 11

Standard deviation ($\sigma$): A measure of variability showing how much scores typically deviate from the mean.

• Standard deviation for male obedience: $\sigma = 3.16$
• Standard deviation for female obedience: $\sigma = 3.68$

These descriptive statistics can be presented in a summary table:

Descriptive measure	Male obedience	Female obedience
Mean perception of obedience	19.7	30.1
Mode obedience score	21	27
Median obedience score	21	30
Range	11	11
Standard deviation ($\sigma$)	3.16	3.68

Graphical representation

Any graphical representation should present data meaningfully while considering the type of data collected. A bar chart effectively illustrates mean scores for perceived obedience, making the difference between male and female perceptions immediately clear to readers.

Analyzing qualitative data

The example practical investigation used a story completion task to gather qualitative data about obedience. Participants completed a scenario about two children, James and Olivia, who were told to tidy their bedrooms by their mother. The first stage of qualitative analysis involves carefully reading all story endings and noting any emerging patterns.

Conducting thematic analysis

Thematic analysis systematically identifies recurring patterns or themes in qualitative data. These themes represent underlying attitudes that participants reveal through their narrative choices.

After reading all story completions, three primary themes emerged: resistance, obedience, and protest. Each theme can be subdivided into more specific sub-themes that capture different manifestations of the broader category. Supporting evidence from participants' stories provides concrete examples of each theme.

Evaluating the practical investigation

Thorough evaluation considers both methodological strengths and limitations of your questionnaire approach.

Strengths

By asking participants about other people's obedience levels rather than their own, the practical investigation reduces social desirability bias. Participants are less likely to display socially desirable responding patterns when discussing others' behaviour. If questions directly asked about participants' own obedience, they might answer dishonestly to present themselves as compliant or independent (depending on what they perceive as socially desirable in their group). Using questions about others' behaviour allows participants to respond honestly, potentially revealing their genuine attitudes. This methodological choice increases the validity of the questionnaire.

The questionnaire in this example gathered 50 responses, representing a quite large sample that can be generalized beyond the immediate participants. This constitutes a strength of questionnaires generally – their ability to be distributed and returned efficiently to large numbers of people. Higher quantities of responses increase confidence in the findings and strengthen conclusions. Though response rates for questionnaires can be low, reducing this advantage when people lack time or motivation to complete them.

Weaknesses

Suggestions for improvement

Since an established questionnaire was not used for this investigation, reliability remains uncertain. Many questionnaires, such as those developed by Adorno et al. (1950) and Cohrs et al. (2012), have undergone extensive testing to confirm their reliability. Checking your questionnaire's reliability would strengthen confidence in the findings.

A test-retest method involves asking the same participants to complete the questionnaire again after a suitable time interval (not too soon to avoid memory effects, but not so long that genuine attitude changes might occur). Comparing scores from both completions would indicate whether the questionnaire consistently measures the same construct. Time-consuming though this approach may be, it provides valuable reliability evidence.

A quicker alternative uses the split-half technique, which divides your questionnaire into two halves and cross-checks the scores obtained in the first half against those in the second half. If scores match, the construct being measured is likely reliable, suggesting your questionnaire consistently assesses the intended concept.

Writing the report

Present your practical investigation following standard reporting conventions used in psychology. These conventions ensure clear, organized communication of your research.

Abstract: Provide a concise summary of the background theory or research, aims, hypotheses, method, results, and discussion. This short paragraph offers an overview of the entire report.

Introduction: Present an overview of related theories and prior research in the topic area. The introduction establishes a rationale for your current investigation by linking previous research to your study aims and hypotheses.

Method: Give a detailed account of your participants, sampling method, apparatus (the questionnaire itself), procedure, controls implemented, and ethical considerations addressed.

Results: Provide a detailed account of the data you gathered and your analysis. Use descriptive statistics and graphical representations to present quantitative findings clearly. Present your thematic analysis with supporting evidence for qualitative data.

Discussion: Draw conclusions from your results analysis, referencing prior research findings. Discuss the strengths and weaknesses of your methodology, evaluating what your investigation achieved and identifying possible improvements for future research.