Experimental Design (AQA AS Psychology)

Independent groups design & random allocation

• There are three main types of experimental design

  • Independent Groups design 

  • Repeated Measures design 

  • Matched Pairs design 

• In an independent groups design participants only experience one condition of the independent variable (IV) e.g.

  • participant A learns a poem with music playing (condition 1)

  • participant B learns the same poem in silence (condition 2)

• Independent groups design generates unrelated data

  • 'Unrelated' refers to the fact that two separate groups are used so that each group generates its own data set

• The performance (scores) of the group in condition 1 is compared to the performance (scores) of the group in condition 2 e.g.

  • participants who have learned a poem with music playing (condition 1) are asked to write down as much of the poem as they can recall in 5 minutes

  • participants who have learned the same poem in silence (condition 2) are asked to write down as much of the poem as they can recall in 5 minutes

  • the dependent variable (DV) is measured as the number of words correctly recalled from the poem

    • Each participant achieves one score from participating in one condition only of the test

• In an independent groups design, participants are randomly allocated to each condition of the IV e.g.

  1. The name of every participant in the sample is put into a hat or a box

  2. The researcher pulls out the first name and allocates this person to condition 1

  3. The researcher pulls out the second name and allocates this person to condition 2

  4. The above procedure continues until all of the participants have been allocated to a condition

• The researcher may use specific name-generator software to randomise allocation to condition, especially if they are dealing with a large sample

• Random allocation to conditions is done to avoid researcher bias  

  • In experiments, researchers manipulate an independent variable to assess its effect on a dependent variable, while controlling for other variables

Evaluation of independent groups design & random allocation

Strengths

• The use of independent groups design means that demand characteristics are unlikely to act as a confounding variable

  • As participants only take part in one condition of the IV they are less likely to guess the aim of the study and act accordingly

  • This increases the internal validity of the study

• As participants experience only one condition it means that order effects are eliminated

  • Participants will not become tired, bored or overly practised at the task

  • This is a strength as it increases the validity of the findings

Limitations

• Participant variables may affect the validity of the findings

  • If more participants with a particular characteristic are all randomly allocated to one condition then this presents an unfair playing field

  • The results are thus not a true measure of the IV's effect on the DV

• More participants are needed for an independent groups design

  • This may cause logistical issues if there are enough people available to take part in the research

  • This means that each condition of the IV may have a lower number than if a repeated measures design was used which in turn affects the reliability of the findings due to the small sample size

Repeated measures design & counterbalancing

• In a repeated measures design participants experience all conditions of the IV e.g.

  • participant A learns a poem with music playing (condition 1)

  • participant A learns a different poem in silence (condition 2)

• Each participant completes each of the experimental conditions

• Repeated measures design generates related data

  • 'Related' refers to the fact that each participant's score on condition 1 is compared to their score on condition 2

    • There is a data set for condition 1 and a data set for condition 2 (in an independent groups design one group represents one condition only)

• Participants act as their own control group as their performance in condition 1 can be compared to their performance in condition 2 e.g.

  • participant A learns a poem with music playing (condition 1) and is asked to write down as much of the poem as they can recall in 5 minutes

  • participant A learns a different poem in silence (condition 2) and is asked to write down as much of the poem as they can recall in 5 minutes

  • the dependent variable (DV) is measured as the number of words correctly recalled from the poem

  • the number of words correctly recalled by the participant in condition 1 (their score for this condition) is compared to the number of words they correctly recalled in condition 2 (their score for this condition)

• A repeated measures design may give rise to order effects which consist of

  • fatigue

    • Taking part in more than one condition may tire the participants which could result in impaired performance in the second condition

  • boredom

    • Spending too long on the experiment may mean that participants lose interest in the experiment which could result in reduced effort from them

  • practice

    • If both conditions of the IV involve a similar sort of task the participants may improve their performance in the second condition

• To avoid order effects researchers use counterbalancing

  • The researcher splits the participants in half e.g. 20 in one group, 20 in the other group

    • Half of the participants experience condition A followed by condition B

    • The other half of the participants experience condition A followed by condition B

Evaluation of repeated measures design & counterbalancing

Strengths

• Participant variables are not an issue with a repeated measures design

  • This is because each participant's performance in one condition is measured against their performance in another condition so they act as their own control group

  • This increases the internal validity of the study

• Fewer participants are needed for a repeated measures design

  • This is because each participant generates two scores it cuts down the need for a larger sample

  • This means that it is less problematic for the researcher to find sufficient participants willing to take part in the research

Limitations

• Demand characteristics may become a confounding variable

  • As participants take part in both conditions of the IV they may guess the aim of the study and act accordingly

  • This decreases the internal validity of the study

• If not controlled for, order effects may lower the validity of the study

  • Participants may become tired, bored or overly practised at the task

  • This is a limitation as the researcher cannot be confident that the IV has affected the DV or that the results were due to other factors

Matched pairs design

• A matched pairs design is one in which participants are matched based on a specific characteristic or variable that is important for the research they are taking part in e.g.

  • age

  • gender

  • ethnicity

  • IQ

  • aggression

• Participants may be matched on more than one variable or characteristic e.g.

  • Maguire et al. (2000) matched their sample of taxi drivers with a sample of controls using the variables of

    • age, gender, handedness (being right-handed)

• By matching participants across conditions, the researcher is assured that one condition does not comprise an over-representation of, for example, males, older people, more aggressive people etc.

• The matched participants are then randomly allocated to one condition each 

• As each participant is related to their pair this design produces related data e.g.

  • in an experiment on the social learning of aggression, participants would be matched on a scale according to how aggressive they already are

  • participant A who ranked 10 for aggression would be matched with participant B who also ranked 10 for aggression 

  • participant A takes part in condition 1 of the experiment; participant B takes part in condition 2 of the experiment

    • This matching helps to factor out aggression as a possible confounding variable in the experiment - any difference in scores should be due to the effect of the IV and not due to natural aggression levels

• Often MZ (Monozygotic/ identical) twins are used in a matched pairs design as they create the perfect matched pair (identical DNA and the same upbringing)

  • One twin can be assigned the experimental condition and the other twin the control condition

Evaluation of matched pairs design

Strengths

• The matched pairs design almost factors out individual differences as a confounding variable

  • Te researcher has worked hard to find a 'match' per participant

  • This means that each participant's performance is compared to someone very similar to themselves so participant variables are controlled to some extent, increasing reliability

• As participants take part in the only condition of the IV this means that demand characteristics are reduced

  • Each participant is tested only once

  • This means that they are less likely to guess the aim of the research which increases the validity of the study

Limitations

• Matching is a difficult and time-consuming process

  • It is often impossible to match the participants across all of the criteria, especially when the unmatched characteristic could be important to the results of the research

  • Even well-matched participants could have different levels of motivation, skill or ability

  • A lack of consistency such as this lowers the reliability of the study

• If one participant drops out of the research then the researcher has to find someone very similar to replace them

  • This is problematic and could slow down the research cycle

  • Therefore funding for the research could be removed if there is a timeline involved