Research Methods Flashcards
Hypothesis
A precise, testable statement about the predicted outcome of the investigation
Directional hypothesis (one tailed)
A prediction which states the direction between two conditions
Non directional hypothesis (two tailed)
A prediction which states that there will be a difference between conditions or that there will be a correlation but which doesn’t state the direction the difference will go in
Extraneous variables
Any variable which may affect the dependent variable other than the IV
Confounding variable
One which affects the dependent variable and which varies systematically with the independent variable
Independent groups
Different participants in each condition of the independent variable
Repeated measures
Each participant does both conditions of the independent variable
Matched pairs
Each person only does one condition of the independent variable but they are matched with another person doing the other condition on some extraneous variable
Counterbalancing in repeated measures
Half the participants do conditions in one particular order and the other half do the condition in the opposite order
This is done to balance possible order effects
Advantages and disadvantages of independent groups
No chance of practise effects between the first time and second time
No chance of participants becoming bored
Participant variables
Advantages and disadvantages of repeated measures
Controls for differences between people
Requires fewer participants
There may be order effects
Boredom
Practise
Lab experimental method
Conducted in a controlled environment
Fired experimental method
Conducted in a real world environment
Natural experiment
The independent variable has not been manipulated by the experimenter but it has changed or occurred naturally
Eg Romanian orphanages
Quasi experiment
The independent variable is not manipulated
Eg investigating differences between men and women on a certain variable
Quantitative data
Information that is gathered that is in numerical form
Negatively skewed distribution
The curve leans over to the right
The mean is less than the median and the mode
Mean, median,mode
Eg test too easy (more students getting high scores)
Positively skewed distribution
Curve leans overs to the left
The mean is more than the median and mode
Mode, median, mean
Eg test too hard
Name the measures of central tendency
Mean
Median
Mode
Name the measures of dispersion
Standard deviation
Interquartile range
Range
Standard deviation
The average about that the scores differ from the mean
The larger the standard deviation, the more the data is spread out
Tells us how consistent people’s scores were
Advantages and disadvantages of standard deviation
Much less affected by outliers compared to the range
Much more sensitive measure of dispersion compared to the range as it uses all the data from the sample in its calculation
Much more difficult to interpret
Name the levels of measurement
Nominal
Ordinal
Interval
Levels of measurement- nominal
Categories
Levels of measurement- ordinal
Rank order
Levels of measurement- interval level
Precise intervals
Safe and unsafe interval scales
Safe- each unit on the scale is exactly the same (eg tape measure)
Unsafe- not directly observable eg depression
We can’t be sure that the distance between each interval is always exactly the same
When to use the sign test
When the DV has been organised into categories (nominal)
When it is a repeated measures of matched pairs design
When we are testing for a difference between conditions
In the sign test, when is the result significant?
If our observed value is equal to or less than the critical value
What does it mean if a result is significant?
It means that is is very unlikely to have occurred by chance
Type one error
You mistakenly conclude that there is an effect when it was just coincidence
To recede the possibility of a type 1 error: choose a stricter level or significance, repeat the study
Type 2 error
You mistakenly conclude there is no effect when there was an effect
To reduce the possibility of a type 2 error: pick a more lenient level of significance
Pilot studies
A small scale trial run which is done before the main study to identify any problems of confusions and to correct these problems.
Random sampling
Every member of the target population has the same chance of appearing in the sample
Opportunity sampling
The researcher uses whoever is available from the target population at the time
Volunteer sampling
Participants chose themselves
Eg replying to a newspaper advertisement
Demand characteristics
The features of a study which may bias the participants to behave in certain ways
Investigator affects
Investigator effects occur when a researcher influences the outcome of research they are conducting. They may have biased the results. For example, if they expect something to happen, this may be reflected through the results (e.g give a certain group of people a higher score)
How to minimise demand characteristics
Testing people without them knowing, using standardised instructions, using a single blind procedure, deceiving people about the true purpose of the study
Reliability
The extent to which findings of measures have been repeated with similar results or the extent to which a measure is consistent
How to test reliability
Inter rater reliability- the observations are made by more than one researcher and the scores between the two researchers are compared to see how similar they are
Test retest reliability- the test is done again on the same sample of participants. We should expect to find a high correlation between the scores on the two different occasions
How to improve reliability
Standardised instructions Train the observers so they mane their observations consistent Operationalise variables Pilot studies Controlled conditions
Validity
The extent to which something accurately investigates what it intends to
Face validity
The extent to which a measure appears on the surface to measure what it is supposed to measure
Concurrent validity
A way of assessing validity by comparing the results with another relevant measure. Eg comparing the results of an IQ test with school results
Ecological validity
The extent to which we can accurately generalise from the results to the real world setting
Temporal validity
The extent to which we can accurately generalise from Feb results of a study that was conducted several years ago to today
How to improve the validity
Control confounding variables
Minimise demand characteristics
Minimise investigator bias
Use objective measures where posible
Primary data
All the data that is gathered by the researchers
Secondary data
Data that wasn’t gathered by the researchers themselves but was gathered for another purpose
Structured interviews
Rigid questions asked in a set order with no variation
Unstructured interview
No set questions decided before the interview
Semi structured interview
Some set questions but time for follow up questions allowed
How to make objective observations
Clearly define the behavioural categories
Train observer thoroughly
Use more than one observer and check inter observer reliability
Case studies
In depth studies of an individual or a small group of individuals regarding a topic of interest to the researcher
Qualitative data
Information collected in non numerical form
Ethical issues
Protection of participant Deception Debriefing Withdrawal Informed consent Confidentiality
Peer review
The scrutiny of research by independent experts
The purpose of peer review
Means that poor quality research shouldn’t be published
Aim
A statement about the purpose of an investigation
What should be included in a debrief?
The aim of the study, the results, the procedure, whether they were deceived, ethical issue considered.
Why are behavioural categories used?
Behavioural categories allow observers to tally observations into operationalised categories. This makes the results easier to analyse
Advantages of open questions in interviews
Lots of information received, more detail
Advantages of closed questions in interviews
Easy to compare specific responses, able to collect and display information easily
Advantages of independent groups
Performances aren’t effected by order effects because each participant only takes part in one condition
Why it is important for research to be replicated
The likelihood of the same differences occurring twice, by chance alone are much smaller than when they occur the first time
Effects that occur in a study are more likely to be reliable if they occur is a repeat of the study- replication increases external reliability
What is a limitation of nominal level of measurement
Does not enable very sensitive analysis
Distinguish between a type 1 and type 2 error
A type 1 error occurs when a researcher claims support for the research hypothesis when they were actually by chance
A type 2 error occurs when the effect the researcher was attempting to demonstrate does exist but the researcher claims there was no significance and accepts the null hypothesis s
In a type 1 error, the null hypothesis is rejected when it is true and in a type 2 error it is retained when it is false
Explain one limitation of a self report technique
Questionable validity, lack objectivity
Discuss how observational research might be improved by conducting observations in a controlled environment
Controlled environment affords the opportunity for control of extraneous variables- could affect the outcome of a study if not controlled
Exclusion of extraneous variables allows for greater inference about cause and effect
Allows the researcher to replicate the observation to check for reliability of the effect.
Explain what is meant by event sampling
Observers/researcher decide on a specific event relevant to the investigation
Relevant event is recorded every time is happens
Explain how using the standard deviation rather than the range would improve a study
Standard deviation is a measure of dispersion that is less easily distorted by a single extreme score (anomaly)
What is a covert observation?
Participants aren’t aware they are being watched
Deception
What is an overt observation?
Participants are aware they are being observed
The observer can be seen by participants
What is content analysis?
A way of indirectly studying concepts in media
What is a pilot study?
- An initial run through of the study
- Check participants know what they are doing
- Check the instructions are clear
- Check the equipment workd
What is concurrent validity?
Comparing a test with another test measuring the same variable.
E.g comparing the results of a maths score with the results of an IQ test
What is counter balancing?
A way of trying to control for order effects in a repeated measures design, e.g. half the participants do condition A followed by B and the other half do B followed by A