All Research Methods Flashcards
What are the two types of self-reporting data?
- Questionnaires
- Interviews
What are Questionnaires?
- Respondents recording their own answers
- Written form
- No face-to-face contact with another person
What are the Strengths for questionnaires (& interviews)?
Genuine/Personal answers
- Respondents reveal more personal answers as they may be anonymous & record the data themselves
Quickly & Easily repeated
- Large numbers of people can do it at the same time → quick info collected
What are the Weakness for questionnaires (& interviews)?
Social Desirability
- Respondents answer questions a certain way that makes them look better as they don’t want to look foolish
Bias
- If all statements in a set of statements are worded favourable/unfavourably → respondents can slip into agreeing/disagreeing with all of them
What type of questions are there?
Open Questions
- Allow the respondent to answer in any way they like - Produce Qualitative data - E.g. Begin questions with ‘How’ or ‘Why’
Closed Questions
Fixed number of possible answers
E.g.
- Yes/No question
- Multiple choice questions - nominal data
- Ranked scales - interval/ratio data
- Likert scale - offers a statement (not a question), nominal
orinterval/ratio level data
- Semantic Differential scale - nominal data or interval/ratio data
What are the strengths for open questions?
Detail
- Provides rich detail answers as it allows people to express their answers freely - Increases validity of data collected
Unexpected findings
- Freely answered - means there can be unexpected findings that closed questions can’t obtain (due to it being limited)
What are the weaknesses for open questions?
Conclusions
- Difficult to draw conclusions because the answers are freely answered & not structured
Subjective
- Interpreting what people mean with the freely expressed answers means that it’s subjective (each researcher having different interpretations)
What are the strengths for closed questions?
Easy analysis - conclusions
- Easy to analyse as the data is Quantitative so conclusions can be easily drawn
Objective
- Objective answers can be made because the limited answers can easily be interpreted the same by researchers
What are the weaknesses for closed questions?
Lack of detail
- Cannot express exact feelings & discover new insights because the researcher determined the choice of answers (limited question) - Data collection low in validity
Oversimplifies
- Oversimplifies human experiences as it suggests that there’s simple answers but respondent may want to express several different views
What type of interviews are there & explain what they are about?
Unstructured Interview
- Loose research aim - Interviewer needs to be skilled at achieving a good rapport - Long and expressive answers - not limited - Gathers qualitative data
Structured Interview
- Standardised - controlled → all respondents are asked the same questions in the same way - Must answer the research aim - Using closed questions - Gather Quantitative data
Semi-structured Interview
- Conversational & dynamic
- Standardised format to follow
- Conversations can flow whilst still achieving the research aim & getting relevant information from respondents
- Gather both Qualitative & Quantitative data
What are Researcher Effects?
- Interviewer characteristics (sex, age, manner & personality) can influence the respondents answer’s
- Important to predict what characteristics might influence respondents & control them
What is a Alternative hypothesis?
There is a relationship between the two variables being studied (one variable has an effect on the other).
What are Sampling Techniques?
- Selecting a small group of participants is a sample
- Unlikely that a whole population can be studied → so sample of the population is gathered using a sampling technique
What different types of sampling techniques are there?
Random Sample
- A sample of participants using a random technique - everyone has an equal chance of being selected
Stratified Sample
- Researcher identifies the different types of people that make up the target population & ensures that every characteristic is represented in the sample
Volunteer Sample
- Self-selecting participants take part (choose to take part themselves rather then being approached or asked by a researcher) by placing an advert in a newspaper
Opportunity Sample
- Asking participants who are available at that time
What are the strengths & weaknesses for these different types of sampling techniques?
Random Sample
- Strengths? - Unbiased - all members of the target population have an equal chance of selection - Weaknesses? - Time-consuming; need to obtain a list of all members in your target population, then identify the sample, then contact the people to ask to take part
Stratified Sample
- Strengths? - Most representative → all subgroups are represented - Specific subgroups chosen according to the important variables (considered by the researcher) → control over any extraneous variables - Weaknesses? - Decision to which subgroup may be used is biased → reducing representativeness of the sample - Long method & participants may not agree to take part→ time-consuming
Volunteer Sample
- Strengths? - Convenient way to find willing participants & are less likely to drop out as they volunteered - Good way to get specialised students for that specific study (putting up an advert near a medical school if the topic is on medical students) - Weaknesses? - Sample biased → volunteer pps have more time in their hands & highly motivated than the population in general - Volunteers may be willing to be more helpful → guessing the aims of the study → demand characteristics
Opportunity Sample
- Strengths?
- Quick & less time → just use the first participants you can find
- Weaknesses?
- Biased → sample is drawn from a small part of the target population → not representative
What types of Data tables are there & what are they about?
Raw data table:
Of all individual values measures in the study
Frequency table:
Shows how many times the scores occurred in a data set
What are the Measures of Central Tendency?
What’s the arithmetic mean?
- It’s calculated by adding up all of the values in a data set & dividing the total by the number of scores collected
What’s the Median?
- Middle value when placed in order
What’s Mode?
- Most frequent score in a data set
What are the Measures of Dispersion?
What’s the Range?
- Difference between highest & lowest
What’s Standard Deviation?
- Distance each score is from the mean
What’s a bar graph?
- Used to present data from a categorical variable, e.g. mean, median or mode
- Placed on the x-axis & the height of the bars represents the value of that variable
- Spaces between bars
What’s a histogram?
- Present the distribution of scores by illustrating the frequency of values in the data set
- No spaces between bars - bars joined together to represent continuous data rather than categorical data
- Values represented on x-axis & height of each bar represents the frequency of the value
What are experiments?
Laboratory Experiment: controlled env
Field Experiment: natural env
Features of experiments:
- Variable manipulated
- Effect can be measured
- Maintain control over other variables
- Set up situation where PPs perform a task
- Performance of task is measured
Exp method:
Theory proposed -> Hypothesis made based on theory -> Variable manipulated (IV) -> Performance measured (DV) -> Theory supported/refuted due to outcome
What type of hypothesis are there?
(One tailed) Directional hypothesis: Direction of results can be predicted
e.g. children will spend longer washing the dishes the more praise they receive
(Two tailed) Non-directional hypothesis: Change or difference is predicted but NOT direction
e.g. praise will affect the time children spend washing dishes
Null Hypothesis: states that there is no difference between groups or no relationship between variables
Experimental/Alternative Hypothesis: Predicting the results of the exp
What types of variables are there?
Independent Variable: Changed by researcher (to see if it causes a change to the dependent variable)
Dependent Variable: Measured, affected by the change of IV
Operationalisation: elaborating what the variables are & how they will be measured
Extraneous variables: Factors that may have an unwanted effect on the DV (that u are not studying)
Confounding variables: Affects study findings directly, that you are no longer measuring what was intended
Situational Variables: Factors in the env affecting exp
Participant Variables: Individual characteristics that affect how a pp responds in exp
What are experimenter effects & demand characteristics?
Experimenter effects: experimenter influence the outcome of an experiment by their actions or presence
Demand characteristics: effect of experimenter causes pps to change their behaviour to meet expectations
What are experimental controls?
Controlling variables that may influence outcomes in exps
Standardisation: Making an exp the same for all pps - highly controlled
Single blind procedure: Pp unaware of the aim of the exp
Double blind procedure: neither the participants nor the experimenters are aware of the aim of the exp
What type of experimental designs are there?
Independent groups: diff people used in each condition
- ppl less likely to guess the aim (don’t know other conditions) -> demand characteristics reduced
- BUT means you need twice as much pps & will be individual characteristics/participant variables between groups (individual diff prevented by random allocation)
Repeated Measures: All pps take part in all conditions
- less pps, reduces individual differences/participant variables
- BUT demand characteristics & order effects increase (practice & fatigue)
Matched Pairs Design: diff Pps assigned to each condition but matched on characteristics
- characters established by background research on pps
- control ensures pps each condition compared fairly
- BUT time-consuming, many pps excluded due to unmatched, difficult to match pps on all characteristics
What are the types of Validity’s?
External validity: How well research findings can be generalised
- Ecological validity: generalised to other situations (daily life)
- population validity: apply to other populations (representative)
Internal validity: Outcome of study is direct result of the manipulated IV
- Construct validity: the test measures what it claims to measure
- Predictive validity: results from a test can predict future behaviour
What is objectivity?
- Need to be judgment free
- DV measured objectively
- Cognitive psychology studies can’t be objectively observed/measured but data can objectively observe data produced by neuroimaging & exps.
What is reliability?
Consistency of findings from research
- Important criterion for scientific
- Exps: re-test reliability is important -> same results again means it’s replicable -> reliable
What is counterbalancing & order effects?
Counterbalancing:
- conditions to minimise the effects of extraneous factors
Order effects:
- order of tasks affect an outcome - lead to fatigue & practice
What’s a normal & skewed distribution & why do we need to examine distribution?
Normal: bell-shaped curve - most of the data in the centre with decreasing amounts evenly distributed to the left & right
- mean, median & mode should be aligned around the mid-point
Skewed: data clumped up on one side or the other with decreasing amounts trailing off to the left and right
e.g.
Positive skew - longer tail on the positive end of values on the horizontal axis, higher scores on a positive skew, higher than the mode.
Negative skew - longer tail on the left side, mean affected by extreme scores by the lower scores on a negative skew, mean will be lower than the mode.
Why? large samples gathered -> useful to examine overall distribution -> show trends that cannot be detected in small samples
Analysis of qualitative data using thematic analysis:
Qualitative data is gathered through interviews, questionnaires, case studies & observations.
What is meant by thematic analysis of this data?
Recording themes, patterns, or trends within data
What’s Inductive & Deductive approach?
Inductive: developing a theory
- Observation
- Observe a pattern
- Develop a theory
Deductive: testing an existing theory
- Start with an existing theory
- Formulate a hypothesis based on existing theory
- Collect data to test hypothesis
- Analyse results
What’s the procedure for thematic analysis?
STAGE 1:
- identifying themes from data that occur frequently/key feature of data
- how frequent to the text the theme is depends on opinions of researchers & nature of material analysed
STAGE 2:
- develop themes into codes -> represents the categories of themes
- use codes to analyse data gathered & search for where it appears in the data
- reviewed & changed if needed - until themes can be supported & used as a summary of the data
THIS FORM OF ANALYSIS IS NON-SCIENTIFIC -> themes are highly dependent on the subjective opinions the researcher -> research bias
What is Correlational Research, including positive, negative & no correlation?
Measuring 2 different variables (co-variables) to see if they are related in any way (plot data on a scatter diagram)
Positive Correlation: Line increases up
- Both co-variables increase
- Perfect positive correlation has a coefficient of +1
Negative Correlation: Line decreases down
- One variable decreases, whilst the other increases
- Perfect negative correlation has a coefficient of -1
No Correlation:
- No clear relationship
- No pattern between variables
- Coefficient of 0
What is the evaluation (pros & cons) for Correlational Research?
Pros:
- Allows looking at a relationship between continuous variables -> determining whether the relationship is significant
- Cost effective -> sometimes use secondary data
Cons:
- Can’t show a cause & effect relationship (which variable influences the other variable)
- May be a 3rd factor affecting data?
What are Inferential Statistics?
- Test of significance tells us if there was effect of IV and DV
- Rest on concept of probability (of data due to being random chance factors)
- Statistical tests tell us if a null hypothesis is true - if results were found by chance then the null must be accepted because no effect has been found. null → no effect.
- If the difference between the data is too small to be significant to show a real effect. the null hypothesis is accepted
- If the result is less than 5% → accept the alternative hypothesis
- If the result is bigger than 5% → accept the null hypothesis
- Calculation of observed value which is then compared to the critical value which shows whether or not the result is real or due to chance.
What is type I & type II error?
Type I:
- Reject the null & choose the alternative
- When probability is too big
Type 2:
- Reject alternative & accept the null
- When probability is too small