Research Methods - Factors Affecting Research Flashcards
(28 cards)
Define validity
The extent which something measures what is supposed to measure. It includes internal validity and external validity.
Internal validity
Refers to whether the results are genuine inside of the study
In the case of an experiment were the effects really caused by the manipulation of the IV or were they caused by other factors?
Love experiments tend of high internal validity because they have high levels of control they tried to control all of the extraneous variables. However, field and natural experiments and non-experimental research methods all have low internal validity due to a lack of control.
External validity
Refers to whether the findings of a study can be generalised to other situations other times and other people outside of the study.
It is ecological validity, the extent to which we can generalise other times and places and population validity the extent to which the results can be generalised to other people.
Ecological validity: research that is carried out in real life setting such as fields and natural experiments tend to have high ecological validity because the initial investigation is already representative of real life. And the result are much more easily generalised to other situations than those of artificial lab experiments.
Population validity: this depends on the sample use in the investigation if there was a large sample which was inclusive of different types of people then the population will be high.
Name the factors which affect internal validity of investigation
Demand characteristics
Social desirability bias
Hawthorn effect
Investigator effects
Extraneous variables
How does demand characteristics affect internal validity?
Cues in the environment help the participant to work out what the research hypothesis is. This may lead to participants altering their natural behaviour to try to go along with the hypothesis. Or they may purposely disrupt the researcher through the screw you effect.
You can control this through
- Deception
- Experimental designs
- Single blind ( participants are told the general aim of the research but are not told which condition they have been placed in)
How does social deseriability bias affect internal validity
Participants usually want to be seen in a good light and therefore may not always give truthful answers to difficult questions their behaviour may also be changed in an attempt to betray themselves more positively.
This can be controlled through
1) deception
2) ensuring confidentiality
How does Hawthorn effect affect internal?
Peoples behaviour will often be affected simply because they are in a study and aware that they are being watched. For some people this will improve performance but other others it might have a negative effect.
How to control this
1) choice of method in some field and natural observations and experiments participants may not be aware that they are being studied
How does investigator effect affect internal validity?
The behaviour and characteristics of the investigator can influence the outcome of the study.
Expectancy affect: the investigator made unknowingly influence the participant to behave in a certain way
Personal characteristics: age ,social class and ethnicity
Mannerisms: smiling ,nodding ,tone of voice
How to deal with this
Avoid contact with the investigator using written instructions
Double blind, research assistance are used to allocate participants to conditions so that neither of the investigator nor the participants know what condition they’re in.
How does extraneous variables affect internal validity?
Anything other than the IV that might have an effect on the results, if something is strong enough to provide a possible alternative explanation for the results it becomes a confounding variable
How to deal with it
1) control, provide a carefully controlled situation
2) sampling technique, try to minimise individual differences between conditions
Name The factors which affect the external validity of investigations
Mundane realism
Sample
Cultural context
How does mundane realism affect external validity?
If a study is too artificial participants will not be behaviour as they would normally. Studies that like mundane realism do not reflect real life and therefore their findings are limited in their value for the real world.
How to deal with it
1) some studies take place in natural real life settings such as field experiment
How does a sample affect external validity?
Samples should be representative of the target population. If findings are too be generalised to people outside of the study. For example, information gained from one person in a case study or from 10 people in a small lab experiment cannot easily be applied to the whole of the population.
How to deal with it
1) Sampling technique there are various ways to select a sample choose a representative one.
How does cultural context affect external validity?
Ideas and funding come from people with particular ideas and opinions, for example most of the money and research is carried out by white American men.
This cannot be fixed and changed, but should be taken into consideration when analysing validity .
Name the methods of overcoming bias in research
1) standardisation, keeping research exactly the same for every participant in the same condition. Control the experiment to the fullest extent. As this limits extraneous variables.
2) randomisation, this means decide in the order of the research using chance not choice. This reduces investigator effects as it means the research cannot affect the outcome knowingly.
Define reliability
The extent to which an investigation is consistent.
Define the types of reliability
Internal reliability, are the measures and procedures within the study consistent.
External reliability, are the results of the study consistent overtime, if repeated would you get the same result? .
How do you test reliability?
Interrater reliability
Split half method
Test retest method
Explain interrater reliability
All observers watch the same behaviour
All observers tally data individually
Compare two or more sets of data
High positive correlation above 0.8 means good reliability
A lower correlation indicates inconsistency .
How to improve interrater reliability
All observers thoroughly trained using video clips
Operationalised definition should be clearly understood
Explain the split half method
It tests internal validity.
Participant answers whole questionnaire
Splits test in half and scores each half separately
Correlate results 0.8 or above indicates consistency
How to improve from split half method
If the correlation is weak, test items may need to be changed
Make sure that the wording of the test items is clear
Make sure that instructions to participants are clear
Explain test retest method
It test reliability overtime (external reliability)
Tests the same participant with the same test on different occasions
Correlate results
0.8 correlation indicates consistency
How to improve test retest
Alter the weight of measuring as it may not be measuring what is intended
Control any variables which may have changed from the 1st to the 2nd to reduce the effects of extraneous variables
Define peer review
Peer review is the assessment of scientific research by others who are experts in the same field.