# AS Psychology Research Methods (2012 JAN EXAM)

All my notes on research methods in preparation for the exam in January 2012.

HideShow resource information

## Experimental Design

Aim: A general statement of what will be investigatd. A good way to start is "To Investigate"

An example of an aim could be: "To investigate whether context is important in memory"

Variables

Independent Variable: The variable manipulated by the researcher (Usually the conditions).

Dependent Variable: The variable measured by the researcher

1 of 18

## Hypotheses

Hypothesis: A prediction or testable statement which we test to see if it is supported or rejected.

Null Hypothesis: States that the IV has no effect on the DV

Experimental Hypothesis: Predicts what will happen in the experiment.

Directional (one tailed) hypothesis)- Predicts the expected direction of the results.

Non-Directional (two tailed) hypothesis) - says that there will be a difference.

2 of 18

## Allocating PPS to conditions...

Independent groups: There are two (or more) conditions and different people are tested in each condition. (e.g. seperate groups.)

Positives: You don't have to worry about order effects.

Negatives: You need a lot of pps - the participant variables aren't controlled.

Repeated Measures: the same participants are tested in two (or more) different conditions.

Positives: Don't need as many participants, and participant variables.

Negatives: May have order effects, and the participants may guess what the research is about.

Matched Pairs: there are different  participants in different  conditions, but they are matched variables which may affect their performance (e.g. IQ, Age, Gender)

Positives: No problem with order effects, and no need to wait between testing.

Negatives: Loss of one person means loss of whole pair. It is also unlikely to be able to control all variables.

3 of 18

## How can we solve the main problems?

Independent Measures:

Problem: Participant variables

Solution: random allocation to conditions.

__________________________________________________

Repeated Designs:

Problem: Order Effects

Solution: Counterbalancing (changing the

order pps do the conditions in)

4 of 18

## Ethical Issues in Psychological Research

What are ethics?: Ethics are moral principles used to guide human behaviour.

• They help us to determine what is considered right and wrong.
• Ethical Issues help protect dignity and rights of participants in the research process.

How do psychologists generally deal with ethical issues?

• Follow the ethical guidelines, these clarify conditions under which research is acceptable.
• The British Psychological code of conduct condemns both research and practice.
5 of 18

## Ethical Comittee

• Decide if a piece or research is acceptable
• Institutions that carry out research have one
• Consider research proposals from perspective of the participants
• Helps correct the power imbalance between the researcher and the participant
• Certain individuals from various disciplines and members of the public.
6 of 18

Punishment: A psychologist who violates the ethical code appears before a comittee.

- The psychologist may be prevented from working in the profession.

Cost - Benefit Analysis:

• This involves weighing up the costs and rewards of a piece of research.
• If the end is for the good of humankind, then we may feel that an undesireable behavior is acceptable.
7 of 18

## Reliability and Validity

Reliable: Means if a study is repeated you would get the same results.

Inter-Rater method: two or more observers rate the behavior and compare the results.

Test-Retest method: Same test is given to pps on 2 separate occasions and compared to see if their scores are similar.

8 of 18

Validity: Valid is whether the results measure what they are supposed to be measuring, so it refers to the accuracy of the data.

Internal Validity: Are the effects observed genuine? (do pps behave naturally? or is their behavior caused by something else?)

External Validity: Extent to which findings can be generalized beyond the experiment. There are three types:

• Population: Generalized to other populations
• Ecological: Generalized to other settings
• Historical: Generalized to past and future
9 of 18

## Improving Validity

• Ensure experimental design is as precise as possible (e.g. use a pilot study)
• The more representative the sample the better
• Moving experiments away from lab settings improves external validity.
10 of 18

## Measures of Dispertion

Range: Difference between the highest and lowest score in any condition, the higher the range, the more spread out your scores.

Advantage: Easy to calculate, takes account of extreme values.

Disadvantage: Greatly influenced by one extreme score.

Standard Deviation: Measures the spread (distribution) of scores around the mean. Calculates difference between each individual score and the mean for the sample.

11 of 18

## What does it tell us?

If a set of scores has a large standard deviation, this indicates a wide distribution (i.e. participants score very differently)

• if it has a small standard deviation, this indicates a narrow distribution.
• It helps s interpret individual scores, based on a normal distribution curve.
12 of 18

• It takes account of all the scores providing a sensitive measure of dispertion.
• Describes the spread of scores within a normal distribution with precision.

• It's harder to work out than the range.
13 of 18

## Correlational Analysis:

What is it?

• It's a statistical technique for analysing quantative data (gained via any method)
• it measures the strength of a relationship between two or more variables.

Types of correlation:

Positive Correlation: The two variables increase together - E.G. the more you smoke, the more chance you have of getting lung cancer.

Negative Correlation - As one variable increases the other decreases - E.G. The more alcohol you drink, the lower your reaction time.

14 of 18

Correlation Coefficient - Is a number between +1 and -1 that expresses the degree to which two variables are related.

Perfect positive correlation is +1.0

Perfect negative correlation is -1.0

15 of 18

A zero correlation is a complete lack of relationship! Scattergraphs are used to show us the correlation between two sets of data.

___________________________________________________

Advantages: Can be used when an experiment is unethical or impractical. And if a correlation is not significant you can rule out a cause and effect relationship.

Disadvantages- Can be misinterpreted. Not possible to establish a cause and effect.

May also be other unknown variables explaining why the results occurred.

16 of 18

## Observation.

Advantages of Observation: You dont need to use textbooks or bring any extra equipment, you dont have to rely on anybody else as its what you see.

There are also different kinds of observation, so you have a choice on what kind you choose to suit your chosen experiment.

17 of 18

## Different Kinds of Observational Techniques

Structured Observations: Where the researcher uses various systems to organize observations, within that there are behavioral categories, and sampling procedures. It is more organized than unstructured.

Unstructured Observations: Where the researcher records all relevant behavior but has no system in comparison to a structured observation.

18 of 18