1. Flashcards
What are the three hypotheses and the definitions?
1) Two-tailed test(Non-directional hypothesis)
- simply states that there will be a difference or relationship
2)One-tailed test(Directional hypothesis)
- states the ‘direction’ of the relationship or difference
3) Null hypothesis (no relationship)
Definition of Operationalisation?
This means that the variables must be clearly defined in terms of how they will be measured
Definition of Extraneous variables and examples?
Any additional, unwanted variables that may ultimately have an impact on DV
Examples: time of day, age of participant, gender, temperature in room, etc
Definition of Confounding variables and examples?
A type of extraneous variables which varies systematically with the IV and these can have significant impact on results so it is difficult to draw conclusions.
Examples: IQ, prior knowledge, fitness levels, injury etc
What are demand characteristics?
When you figure out the purpose of the study and participants change their response to either aid or purposefully ruin the research.
What are investigator effects?
Any effects of the investigator’s behaviour (conscious or unconscious) on the research outcome (DV). This includes clues given to the participants making them realise what the research is about or for
What is randomisation?
The use of chance in order to control for the effects of researchers unconscious bias when designing materials and deciding the order of conditions.
What is standardisation?
All participants should be subject to the same environment, information and experience. It includes standarised instructions and keeping the same standard of procedure
What is a ‘true’ experiment?
- random allocation ( less bias)
- must have a control group
- must have manipulation of IV and measurement of DV
Extraneous variables can be subdivided into what two sub-variables?
- Participant variables( any individual differences between participants that may affect the DV)
- Situational variables ( any features of the experimental situation that may affect the DV)
What is a Laboratory Experiment?
It is usually a lab experiment that takes place under controlled conditions, where the researcher manipulates an IV and measures a DV and any extraneous variables are removed and the pps know they are being studied
What is a Field experiment?
An experiment that takes place in a natural setting within which the researcher manipulates the IV and records the effect on the DV. However the measures are less controlled and pps may not know they are being observed.
What is a Natural Experiment?
A natural experiment takes place in a setting where the change is the IV is not manipulated by the researcher, but occurs naturally. The research can control how much exposure the pps gets to the IV and the pps will know they are being investigated.
What is a Quasi Experimental?
It is when the IV has not been decided by anybody but occurs naturally. The IV exists on its own. A quasi experiment lacks some of the key ingredients of a ‘true experiment’. There is high control over variables like gender, age and height.
Evaluate a Lab experiment
+ high level of control over extraneous & confounding variables
+ high level of internal validity
+ replication possible, which increases reliability
- low levels of ecological & external validity
- generalisibality issues
- demand characteristics
- lower levels of mundane realism
- costly
Evaluate a Field Experiment
+ high mundane realism- valid and authentic behaviour
+ high ecological behaviour
+ less likely to be demand characteristics
- lack of control of variables leading to low internal validity
- precise replication difficult
- ethical issues such as consent and privacy
Evaluate a Natural Experiment
+ opportunists for research
+ high external validity
- cannot be randomly allocated
- lacks internal validity
- time consuming waiting for IV to occur
Evaluate a Quasi Experiment
+ high controlled
+replication
+ internal validity
- no random allocation
- confounding variables
- lower internal validity as IV is not manipulated
What is Internal & External Validity?
The extent to which the researcher has measured what they intended to.
This is the degree to which findings can be generalised to other groups of people, and over time.
What are the three types of External Validity?
POPULATION- how well the results can be applied to all populations
ECOLOGICAL- how well results from a test can be generalised to other environments
TEMPORAL- how well results form a test can be applied across time periods
What is Mundane Realism?
Degree to which the task used to assess behaviour reflects activities that would happen in every day life.
What are the three experimental designs?
Independent groups
Repeated measures
Matched pairs
Explain the concept of Independent groups
- two separate groups of pps experience a different level of the IV
- used when you want to investigate a difference BETWEEN groups
Explain the concept of Repeated measures
- all pps experience BOTH conditions
- used when investigating a difference WITHIN groups
Explain the concept of Matched pairs
- pps are matched on areas related to the study in which they are similar ( IQ, age , ability etc)
- attempts to control confounding and participant variables
Evaluate the INDEPENDENT GROUPS
+ pps less likely to guess the aim
+ order effects less likely
+ less time consuming if conditions can be carried out simultaneously
- reduced validity
- less economical
- twice as many pps required
- confounding variables
Evaluate the REPEATED MEASURES
+ less time consuming to recruit
+ no participant variables
+ internal validity increase
- more likely to guess the aim- demand characteristics (socially desirable thoughts)
- effects may carry on to the second condition
Evaluate the MATCHED PAIRS
+ order effects and demand characteristics less likely as PPS only take part in once condition
+ matching process reduces pps variables
- pps can never be matched directly
- matching is time consuming and expensive
What are the 6 types of OBSERVATION? Evaluate observations.
-naturalistic
-controlled
-covert
-overt
-participant
-non-participant
OBSERVATIONS are good for capturing what people actually do but have observer bias
naturalistic vs controlled observation
NATURALISTIC: when observations are done in a natural ordeal-life settings(hospitals, homes, schools etc.) The observer makes no effort to control or manipulate the situation for making an observation.
CONTROLLED: when you control certain factors that determine behaviour as they are not the focus of your study(laboratory) eg. smoke could only be introduced in a controlled laboratory experiment (Bibb Latane and John Darley conducted a study in 1970)
covert vs overt
COVERT: observations take place without obtaining consent first. PPS are unaware they are being observed.
OVERT: observations take place with consent and pps are aware they are being watched
participant vs non-participant observation
Participant: Researcher is active member of the group being observed
Non-Participant: Researchers try to conceal their presence while making observations and are separate from the group
Evaluate naturalistic vs controlled
NATURALISTIC: + high external validity as findings can be generalised to every day life
- lack of control makes replication difficult
CONTROLLED: + confounding/extraneous variables may be less of a factor so replication of the observation becomes easier
- may produce findings that cannot be readily applied to everyday life
Evaluate covert vs overt
COVERT : + removes demand characteristics
+ increase internal validity
- ethics may be questioned as people
OVERT: + more ethically acceptable
- demand characteristics may have some influence on behaviour
Evaluate participant vs non-participant
PARTICIPANT: + increased insight into people lives
+ increase external validity
- lose objectivity( as researchers can be going native)
NON-PARTICPANT: + researcher maintains objective psychological distance
- may lose valuable insight as theirs distance
What is a unstructured observation and strengths and weaknesses?
- You write down everything you observe
- better when sample size is small and likelihood of producing large amounts of data is less
+ rich in detail - observer bias
- difficult and time consuming to analyse
What is a structured observation and strengths and weaknesses?
- researcher makes behavioural categories and observe and only write down info that fits into those categories
- better when there is too much target behaviour occurring and its difficult to record
+ less susceptible to bias
+ data easier and less time consuming to analyse - lacking in meaningful depth data
What should behavioural categories be?
- distinct
- definable
- measurable
What are the types of sampling methods?
1) time sampling - recording behaviour in predetermined timeframes.
2) event sampling- counting the number of times a specific ‘event’ occurs.
Evaluate both types of sampling methods?
Time: + reduces total number of observations needed
- behaviour may be unrepresentative of general or target population
Event: + good for infrequent behaviour which may be missed when using time sampling
- research may overlook behaviours if categories are complex or confusing
Within questionnaires what are the two type of questions?
OPEN: no fixed range on answers, with wide range of different responses and qualitative data
CLOSED: fixed set of responses so yes or no , maybe rate 1-10 and it’s quantitative data
Different types of interviews and what is interview schedules?
STURCUTRED: strict set of questions to be asked in the same manner and order
SEMI-STRUCTURED: predetermined questions with follow-up questions
UNSTRUCTURED: no set questions so like a convo
Interview schedules are standardised set of questions that interviews will follow with aims to reduce interviewer bias and questions must not be leading but probes can be used
What are the styles of closed questions?
Likert: the respondent indicates the degree in which they agree with a statement. The scale would usually range from strongly agree to strongly disagree
Rank: similar to LIKERT but pps would identify a numerical value that represents their feelings towards the topic
Fixed-choice: choice of possible options that they can choose from and will be told to tick ‘ all that apply’
Evaluate questionnaires and Interviews
QUESTIONNAIRES: + cost-effective
+ large amounts of data gathered quickly
+ no need for a researcher present so reduces effort
+ straightforward to analyse in forms statistical analysis
- responses may not be truthful
- demand characteristic called social desirability bias
- response bias as respondents answer in the same way
- might be rushed
INTERVIEWS: + structured are easy to replicate
+ more flexibility in unstructured
+ unstructured should be more truthful responses as they are detailed reponses
- limits the richness of data as theirs a nature of the structured interview
- increased risk of interview bias as unstructured can be made up questions
- harder to analyse as there could be irrelevant info
- unstructured may lie for reasons of social desirability
How does a correlation differ from an experiment?
- an experiment controls and manipulates the IV in order to measure the DV. This deliberate change allows us to infer that IV has a causal effect on the DV
- in a correlation there is no manipulation of an IV so it is not possible to establish a cause and effect relationship
Evaluate correlations
+ preliminary research, saves time no experiment needed
+ more ethical- no changing anything
+ clear as correlation signifies a clear relationship or absence there of
- causation as we cant determine causes of correlation
- we don’t know why there’s a correlation
- can only draw inferences so lacks internal/external validity
- third variable so extraneous variables
Evaluate Case studies
STRENGTH: Rich detail-insight into unusual/atypical behaviour-valid
STRENGTH: Contribute to understanding of normal function-HM (separate memory stores)
STRENGTH: Can generate hypotheses for future study.
LIMITATION: Small sample- can’t generalise findings.
LIMITATION: Information based on subjective selection by researcher.
LIMITATION: Personal accounts- inaccurate, memory decay.
LIMITATION: Low validity.
What are the types of ethical issues?
informed consent, deception, protection from harm, privacy and confidentiality, right to withdraw, debrief.
(cost benefit analysis is looked into to decide whether a prospect can go ahead)
What are the three types of informed consent?
PRESUMPTIVE- similar group of people asked if the study is acceptable, if answer is yes, presumptive consent from the original group is decided
PRIOR GENERAL- pps give permission to take part in a number of studies, one of which may involve deception
RETROSPECTIVE- pps give consent after they have already taken part during a debrief
What are the different sampling methods?
Random- each person in the target population has an equal chance of taking part
Volunteer- the study id advertised to the population
Opportunity- researcher asks anyone who is close by
Systematic- every nth member is selected
Stratified- the sample reflects the proportions of sub groups and groups that make up the population are identified.
Evaluate random sampling techniques
+ truly random
+ less subjective to bias
+ more likely to achieve a representative sample ( better external validity)
- difficult and time consuming to obtain a full list of the population
- results could still be distorted
- could result in a volunteer sample
Evaluate volunteer sampling technique
+ easy and require minimal time and effort to obtain a sample
+ pps will be more engaged with the research than people off the street
- volunteer profile bias
- external validity
- ulterior motives
- internal validity
Evaluate opportunity sampling technique
+ convenient and less time consuming
+ less costly
- more likely to get an unrepresentative sample so low external validity
- could be subject to researcher bias
Evaluate systematic sampling technique
+ objective
+ no researcher bias
- time consuming
- may result in a volunteer sample
Evaluate stratified sampling technique
+ sample is representative so high external validity
+ generalisation is possible
- cannot identify all the ways people are different
- time consuming
What is a pilot study and how do we conduct one?
- its a small-scale test of the methods and procedures to be used on a larger scale to assess the acceptability of an approach to be used in a large-scale study
- may involve a handful of pps in order to ‘ road test’ the procedure and check the investigation runs smoothly.
What is content analysis?
observational study of people through their indirect communication media so it can be diaries, interview transcripts and status updates. Analyse and turn the qualitative data into quantitative data to allow it to become more objective and identify trends.
What is thematic analysis?
the analysis of patterns and themes within data and analysis of qualitative data usually is a transcript form, such as an interview or open questionnaire. Similar to content analysis in that data is coded ( it can’t be a video) and themes are developed( recurring idea)
Explain the method on content analysis?
1) data is collected
2) the researcher familarises themselves with the data
3) the researcher identifies coding units( categories)
4) the researcher applies the coding units, tallying how many times they occur
Explain the method of thematic analysis?
1) familarisation of data
2)generate initial codes
3) search for themes
4) review themes
5) define and name themes
6) write a report
Evaluate content analysis
STRENGTHS: - quick and easy to carry out
- allows for statistical analysis
- easily replicated by others
- based on observations of what people actually do
- the person who created the artefact didn’t know it could be used for content analysis
WEAKNESSES:- causality not established
- could be culturally or age biased
- can be ‘time-locked’
- data is merely described
- interpretation of coding might be different
Within descriptive statistics what are measures of dispersion
1) standard deviation- dispersion of data around the mean
2) range- dispersions of a data set of scores
Within descriptive statistics what are measures of central tendency
1) mode- most frequently occurring and used when data is in categories and is not affected by outliers and is a very crude measure
2) median- middle number and easy to calculate and extreme scores do not affect the median
3) mean- the average number and its the most sensitive of measures as it takes all values in data set into account, so it is most representative measure however it can be distorted by extreme values( anomalies)
Within Data Representation what are bar charts used for?
- they allow for categorical data to be compared and presented visually BUT THEY SHOULD NOT TOUCH
- used when there’s discrete data when there are two conditions, IV on x-axis, DV along the y-axis
Within Data Representation what are histograms used for?
- displays continuous data and BARS DO NOT TOUCH
- x-axis has the IV and y-axis has the frequency within each interval
- need to figure out class width and frequency divides by class gives frequency density
What are Scattergraphs used for?
Shows the relationship between two variables ( dependent variable)
What are the three types of distrubution graphs
Distribution graphs give us a virtual how data is spread out. Depending on the direction of the distribution, we can infer where the majority of the data sits.
1) NEGATIVE SKEW: mode is at the highest peak, median in the middle and mean is dragged towards the tail
2) NORMAL DISTRIBUTION: mean, median and mode would all fall in the centre of the bell curve
3) POSITIVE SKEW: mode is at the highest peak, median in the middle and mean is dragged towards the tail
What are the three types of data within levels of measurement?
Give an explanation and and example
1) NOMINAL: - data that can be places in categories specifically tables or bar charts
- you’re able to count the number of items in a category
e.g. count how many people prefer maths or english
2) ORDINAL: ranked or ordered numerical data so its an opinion - does not have equal intervals
e.g. ranking anxiety, happiness and enjoyment of football
3) INTERVAL: - numerical data - doesn’t have equal intervals
- universal intervals
e.g. anything that increases or decreases in precise increments such as temperature, weight and height
What are the ways of assessing reliability?
- Experiments: replication (if the results are the same or very similar, the experiment is reliable )
-Self-report( can be surveys or interviews): test-retest reliability - Observations: inter-observer reliability
What is test-retest reliability?
- the same test is administered to the same person twice at different times.
- if the test is reliable, the results will be the same or at the least very close across all tests
- scores are correlated to show whether similar, a coefficient +0.8 or higher shows a significant similarity, so test is reliable
What are the ways to improve reliability in questionnaires, interviews, observations and experiments?
QUESTIONNAIRES : produces low test- retest reliability may require some of the items to be ‘deselected’ or rewritten.
INTERVIEWS : use the same interviews each time and interviews must be properly trained and use a structured interviews.
OBSERVATIONS : make sure behavioural categories have been properly opertionalised, measurable and self-evident. Categories should not overlap and all possible behaviours should be covered on the checklist and observer may need further training in using the behavioural categories
EXPERIMENTS : use standarised procedure for types of environments and the way you recruit pps. Procedures must be consistent everytime.
What are the two ways to assess validity?
Face validity : a basic form of validity in which a measure is scrutinised to determine to determine whether it appears to measure what it is supposed to measure
Concurrent validity : a form of testing validity whereby pps receive a new measure against an already valid measure to check whether the results are similar.
What are the ways to improve validity in questionnaires, observations and experiments?
EXPERIMENTS : redesign procedure or test to make it more realistic to improve mundane realism and clearer method. Better opertionalisation of variables , control variables. Replication lab experiments in a more natural setting to see if results compare and make sure your sample is representative- age, gender and intelligence.
OBSERVATIONS: better operationalisation of behavioural checklist and change the types of sampling( time/ event) , retrain observers and conduct in other settings, times , data and change the pps way of being sampled ( volunteer, stratified…)
QUESTIONNARIES: revise questions ad re-check concurrent validity.
What is the point of Peer Review?
- involves aspects of the written investigation being scrutinised by a small group of usually two or three experts in the particular field. These experts should conduct an objective review and be unknown to the author or researcher
What are the stages of peer review process?
1) to allocate research funding
2) to validate the quality and relevance of research ( assessing quality and accuracy for hypotheses, methods, tests and conclusions drawn)
2) to suggest amendments or improvements ( in extreme cases this means concluding work inappropriate for publication or should be withdrawn)
Evaluation of Peer Review
- Anonymity, researcher likely to produce honest appraisal but a minority use this as a way of criticising rivals
- Publication bias , tendency for editors of journalists to publish significant headline grabbing findings to increase credibility. Research that doesn’t show positive results may be discarded creating a false impression of psych and researchers could self review which could be very biased
-Process may suppress opposition to mainstream theories wishing to maintain the status quo. Reviewers tend to be more critical of research that contradicts their own view.
Within Data Analysis what are the kinds of Data?
Qualitative data- data in words but can be converted to numbers for the purpose of analysis
Quantitative data- data that usually given as numbers
Primary data - info that has been obtained first-hand by the researcher for the purposes of a research project
Secondary Data- info collected by someone else and so pre-dates the current research project
Evaluate Qualitative Data
+ can offer much more richness of detail than quantitative data
+ tends to have greater external validity as more meaningful insight into the pps worldview
- more difficult to analyse as its hard to be summarised statistically
- conclusions often rely on the subjective interpretations of the researcher and these may be subject to bias
Evaluate Quantitative Data
+ relatively simple to analyse , therefore comparisons between groups can be easily drawn
+ data in numerical form tend to be more objective and less open to bias
- its too much narrower in scope and meaning than qualitative data. Therefore may fail to represent ‘real-life’
Evaluate Primary Data
+ it fits the job so authentic data obtained is specific info that the researcher requires
- requires time and effort on the part of the researcher.
Evaluate Secondary Data
+ unlike primary data, secondary may be inexpensive and easily accessed requiring minimal effort
+ no need to conduct primary data collection so time efficient
- there may be substantial variation in the quality and accuracy of secondary data
- content of data may not match researcher’s needs or objective
Within Inferential Statistics what is the need for stat learning?
- provides a way of determining whether hypotheses should be accepted or rejected. In psychology, they tell us whether differences or relationships between variables are statistically significant or have occurred by chance.
Within Inferential Statistics what is probability and the accepted level?
- all studies employ what we call a significance level in order for significant differences in relationships. The accepted level is 0.05 (5%). This is the level at which the researcher decides to accept the research/ alternative hypothesis or not.
Why is 0.05 (5%) level of significance used?
- its because it means there will be no danger to human life
- it also allows us to balance type 1 & 11 errors
What is the Critical Value?
A single number found from a table of critical values, which is compared to observed / calculated value to work out if the results of the study are significant.
What is a Type 1 Error?
Rejecting null hypothesis and the alternative hypothesis is wrongly accepted. Often referred to as an optimistic error or false positive as there claims to be significant effect when there is not. This is usually occurs when the level is more strict ( stringent- 0.01 instead of 0.05)
What is a Type 11 Error?
Accepting null hypothesis and the alternative hypothesis is wrongly rejected. Often referred to as a pessimistic error or a false negative. There is no relationship. More commonly made when the significance level is too high ( too lenient ) such as 0.1 rather than 0.05 / 5% as significance values could be missed
Reporting psychological investigations - Abstract
Its the first section in a journal (150-200 words)
Summary of report, aim, hypothesis, method, results & conclusions. Psychologists read these when researching to see if the investigations are worth further examination
Reporting psychological investigations - Introduction
Literature review of the general area of investigation detailing relevant theories , concepts and studies that are relevant to the study. Follows a logical progression.
Reporting psychological investigations - Method
Needs to include sufficient detail so others can replicate it in these sub sections: Replication, design, sample, materials, procedure, ethics
Reporting psychological investigations - Results
Needs to summarise Key findings
descriptive statistics (graphs, tables etc.), measures central tendency & measures of dispersion, Inferential statistics (statistical test, calculated & critical value, level of significance)
Reporting psychological investigations - Discussion
Summarise results in verbal form. Should be discussed in the context of the evidence presented in intro and should mindful of the limitations of the present investigation. Might include wider implications of the research like RWA and what contribution this investigation has made to existing knowledge within the field.
Reporting psychological investigations - Referencing
Full details of any source material that the researcher drew upon or cited in the report. May include journal articles,books and websites
When choosing a stats test what are the three things to consider?
1)whether a researcher is looking for a difference or a correlation/ association
2) If a difference what experimental design is being used- independent or group or repeated measures
3) the level of measurement for the data - nominal, ordinal or interval
What is a binomial sign test? How do we calculate one?
-tests if there is a difference between the data collected in 2 conditions so for example to compare a pps score when they have been tested in 2 conditions.
-needs RMD and nominal data
1) Work out the ‘calculated value’ (s) for your data set
2) Find N this is the number of differences in our data set
3) Compare your ‘s’ to the critical value
4) If the calculated values is LESS than or EQUAL to the critical value you have a significant results and can REJECT the NULL hypothesis and ACCEPT EXPERIMENTAL hypothesis.
VICE VERSA
What is the Paradigm War? What is Kuhn 1962 Paradigms approach ?
This is how psychology is being accepted as a ‘normal’ science and accepting the principles.
Suggested that what distinguishes scientific disciplines from non-scientific is a shared set of assumptions and methods. Lack a universally accepted paradigm and natural sciences have a set paradigms.
The Hypothetic - deductive method (Popper)
1) A theory is conducted
2) Hypothesis - how is it tested
3) Empirical data - are findings replicable?
4) Interpretation - can findings be generalised and how is the theory altered based on this.
Karl Popper 1934’s argument on Falsifiability
- argued that the key criteria of a scientific theory is its falsifiability. That sciences need to be able to be proven false even if its not false which makes them stronger theories.
Karl Popper 1934’s argument on Replicability and Objectivity and the empirical method?
- replicability of findings over a number of times increases reliability. This demonstrates that findings are precise over a number of circumstances which aids scientific rigour.
- scientific researchers must strive to maintain objectivity by keeping critical distance and not allowing personal biases to discolour data or influence behaviour of pps
- objectivity is the basis of the empirical method and it emphasises the importance of the data collection based on direct, sensory experience.
Acronym to remember all of the tests
Carrots Should Come Mashed With Swede Under Roast Potatoes.
