Research Methods Flashcards
Research methods
The means by which explanations are tested
Experimental method
A research method method using random allocation of participants and the manipulation of variables to determine cause and effect
Independent variable (IV)
The factor manipulated by researchers in an investigation
Dependent variable (DV)
The factor measured by the researcher in an investigation
Operationalisation of variables
The process of defining variables into measurable factors
Extraneous variables (EV)
Variables other than the IV that might affect the DV
Extraneous variable factors
- Guessing the purpose of research and trying to please researcher by giving the ‘right results
- “”trying to annoy the researcher by giving the wrong results called the screw you effect
- Acting unnaturally out of nervousness or fear if evaluation
- Acting unnaturally due to social desirability bias
Single blind procedure
Teaching that reduces demand characteristics, involves ppts having no idea which condition of a study they are in
Participant variables
Concerns factors such as ppts age and intelligence
Situational variables
Concern the experimental setting and surrounding environment eg temp and noise levels
Experimenter variable
Concerns in the personality, appearance and conduct of the researcher, eg female researcher may gain different results to a male one
Confounding variables
Uncontrolled extraneous variables that negatively affect results
Control
Random allocation and counterbalancing, randomisation and standardisation
Random allocation
All individuals in sample have equal chance of getting picked, decreases systematic error so individual difference in responses/ ability are far less likely to consistently affect results
Counterbalancing
Method used to deal with extraneous effects when using repeated measures design
ABBA
Half do Condition A followed by Condition B
Other half Condition B followed by Condition A
Order effects BALANCED OUT by the opposing half of ppts
Randomisation
Used in presentation of trails to avoid any systematic error that the order of the trails might present
Standardisation
Refers to the process in which procedures used in research are kept the same
Demand characteristic
Features of a piece of reseats which allows an ppt to work out aim or hypothesis. Ppts May then change their behaviour and so frustrate the aim of the research
Investigator effects
A researcher effect where researcher influence ppts response
Factors affecting investigator effects
Physical characteristics- age or ethnicity
Less obvious personal characteristics- tone of voice
Investigator may be unconsciously bias in their interpretation of data and find what they expect to find
Double blind
Double blind
Procedure to reduce investigator effects, neither ppts or investigator knows what condition ppts are in
Laboratory experiment
Experiment conducted in a control environment allowing the establishment of causality
Adv and Dis of laboratory experiment
Adv High degree of control Replication Cause and effect Isolation of variables
Dis Experimenter bias Problems operationalising the IV and DV Low external (ecological) validity Demand characteristics
Field experiment
Experiment conducted in a naturalistic environment where the researchers manipulate the independent variable
Natural experiment
An experiment where the independent variable varies naturally
Quasi experiment
Where the researcher is unable to freely manipulate the independent variable or randomly allocate ppts to the different conditions
Used when the researcher is interested in independent variables that cannot be randomly assigned
Adv and Dis of field and natural experiments
Adv
High ecological validity
No demand characteristics
Dis Less control Replication Ethics Sample bias
Observational techniques
Involves watching and recording behaviour
Most observations are naturalistic (occurs in real world setting)
Control conditions eg Milgrams 1963
Types of observations
Participant observation
-observed become activity involved in situation being studied to gain a more hands on perspective eg Zimbardo’s
Non participant observations
-involves researchers not become actively involved in the behaviour being studied eg Ainsworth strange situation
Overt and covert
Overt - where participants are aware they are being observed eg zimbardo 1971
Covert - where participants remain unaware of being observed eg festinger 1957) study where he infiltrated a cult who were prophesying the end of the world
Adv and dis of observational techniques
Adv
High external validity
Practical method
Few demand characters
Dis Cause and effect Observer bias Replication Ethics Practical problems
Sampling procedures
Event sampling - counting the number of times a behaviour occurs in a target individual/s
Time sampling - counting behaviour in a set time frame eg recording what behaviour is being exhibited every 30 seconds
Behaviour categories
When conducting stricter observations, psychologists have to decide which specific behaviours should be examined, operationalise the behaviour through the use of behaviour categories
Inter rater reliability
Where observers consistently code behaviour in the same way
Self report techniques
Participants giving info about themselves without researcher interference
Questionnaire
Self report method where ppt record their own answers to a pre set list of questions
Questionnaires - types of questions
Closed (fixed questions) - involves yes/no answers - easy to quantity but restrict answers
Open questions - allow ppts to answer in their own words. There are more difficult to analyse but allow freedom of expression and greater depth of answers
Adv and Dis Questionnaires
Adv Quick Lack of investigator effects Quantitative and qualitative analysis Replication
Dis Misunderstanding Biased sample Low response rate Superficial issues Social desirability
Questionnaire construction
Aim Length Previous questions Pilot study Measurement scale
Interviews
Self report method where participants answer questions in face to face situations
Types of interviews
Structured - identical closed questions, interviewers do not need much training
Unstructured - informal discussion on a particular topic, asking follow up questions. Interviewers need training and skill
Semi structured - combining both producing quantitative and qualitative
Adv and Dis interviews
Adv Complex issues Waste misunderstandings Data analysis Replication
Dis Interviews effects Interview training Ethical issues Participant answers
Design of interviews
Gender and age
Ethnicity
Personal characteristics and adapted role
Correlational studies
The factors measured in a correlational study to assess their direction and strength of relationships
Co- variable
The variables investigated in a correlation
Positive n negative correlation
P- occurs when one co variable increases as another co variable increases
N- when one co variable increases with another co variable decreases
Scattergrams
Type of graphs used to display the extent to which two variables are correlated
Adv and Dis of correlational analysis
Strength
Allows predictions to be made
Allows quantification of relationships
No manipulation
Weaknesses Quantification problem Cause and effect Extraneous relationships Only works for linear relationships
Case studies
In depth detailed investigation of one individual or small group
Case studies Adv and Dis
Adv
Rich detail
Sometimes only possible method
Useful for theory contraction
Dis
Not representative
Researcher bias
Reliance on memory
Difference between experiment and correlation
E- isolated and maniples IV to observe its effect on DV and controls environment (EV) Establish cause and effect
C- identified variables and looks for a relationship between them
Scientific process
Aim is precise statement of why a study is taking place
Hypothesis is a precise, testable prediction
The experimental/ alternative hypothesis
Predicts that differences in the DV will be beyond the boundaries of chance ( they will occur as a result of manipulation of IV)
The null hypothesis
Is ‘the hypothesis of no differences’ predicts that he IV will not affect the DV
Directional (one tailed) hypothesis
Predicts the direction of the results eg significant reduction in speed of reaction times as a result of caffeine consumption
Non directional (two tailed)
Predicts there will be a difference but does not predict the direction of the results
Example of hypothesis
There is a difference in number of verbal errors made by ppts who believe there are 5 listeners (small audience) and by ppts who believe there are 100 listeners (large audience)
Sampling
The selection of ppts to represent a wider population
Population vs sample
Main difference is to do with how observations are assigned to the data set.
A population includes all of the events from a set of data.
A sample consists of one or more observations drawn from the population
Random sampling
Where each member of a population has an equal chance of being selected
Adv and Dis random sampling
Adv Unbiased selection Generalisation Dis Impractical Not representative
Opportunity sampling
Involves selecting ppts who are available and willing to take part, for example asking people in the street who are passing
Adv and Dis opportunity sampling
Adv
Ease of formation
Natural experiments
Dis
Unrepresentative
Self selection
Volunteer sampling
People volunteering to participate
Adv and Dis volunteer sampling
Adv
Ease of formation
Less chance of screw you effect
Dis
Unrepresentative
Demand characteristics
Systematic sampling
Involves taking every nth person from a list to create a sample
Adv and Dis of systematic sampling
Adv
Unbiased selection
Generalisation
Dis
Periodic traits
Not representative
Stratified sampling
Is a small scale reproduction. Involves dividing a population into characteristics important for research eg age social class
Adv and Dis of stratified sample
Adv
Representative
Unbiased
Dis
Knowledge of population characteristics requires
Time consuming
Pilot studies
Small scale practice investigation
Examine the feasibility of an approach that is intended to ultimately be used in a larger scale study
Experimental designs
Independent group design
Repeated measures design
Matched pairs design
Independent group design
Experimental design in which each participant performs one condition of an experiment
Adv and Dis independent group design
Adv
No order effects
Demand characteristics
Time saved
Dis
More participants needed
Group differences
Repeated measure design
Experimental design where each ppt performs all conditions of an experiment
Adv and Dis Repeated measure design
Adv
Group differences
More data/ fewer participants
Dis Order effects Counterbalancing Demand characteristics Takes more time
Matched pairs design
Experimental design where ppt are in similar pairs with one of each pair performing each condition
Adv and Dis Matched pairs
Adv
No order effects
Demand characteristics
Group differences
Dis
More participants
Matching is difficult
Time consuming
Ethical issues
The rules governing the conduct of researchers in investigations
Code of ethics include
Can Do Can’t Do With Participants Informed consent Deception Confidentially Debriefing Withdrawal Protection from harm
Observation research
Observations are only made in public places where people might expect to be observed by strangers
Incentives to take part
Participants should not be offered bribes or promised readers for their participant as this puts pressure on them to take part
Peer reviews
Process takes place before a study is published to ensure that research is of high quality, contributes to the field of research and is accurately presented
Without this, poor research might be disseminated which would damage the integrity of that field of research or that of the disciple as a whole
Adv and Dis of peer review
Adv Maintains high standards in research Helps prevent scientific fraud Dis Experts with conflict of interest may not approve finding to save own repulsion File drawer effect
Implications of psychology research for the economy
Practical application
Effective therapies, developed through research, making huge savings in financial costs, allowing many people to return to work and contribute more fully to the economy
Evaluation of psychologist research for economy
Reduces costs eg psychologically healthy people less likely to incur costs of health services (risk factors)
Psychologists need to be aware that ethical consideration comes before profit, research should no be used to exploit people as this has negative consequences
Reliability
The extent to which a test or measurements produces consistent results
Types of reliability
Internal reliability
External reliability
Internal reliability
Concerns the extent to which something is consistent within itself eg set of scales should measure same weigh between 50-100g as between 150-200g
External reliability
Concerns the extent to which a test measures consistently over time
Ways of assessing reliability
Split half method - measures internal reliability but splitting a test into toe and having the same ppt do both halves. If the two halves of the test provide similar results then this indicates the test has internal reliability
Test retest method - measures external validity, giving the same test to the same ppts on two occasions, if same results obtained then reliability is established
Inter rater reliability - is a mean of assessing whether different observers are viewing and rating behaviours the same way
Validity
The extent to which results accurately measure what they’re supposed to measure
Types of validity
Internal validity
External validity
Internal validity
Concerns whether results are due to manipulation of the IV and have not been affect led by confounding variables - improved by reducing investigator effects, minimising demand characteristics, standardised instructors and random samples
External validity
Refers to the extent to which an experimental effect (the results) can be generalised to other settings (ecological validity) other people (population validity) and over time (temporal validity)
Eg milgrams study lacked external validity
Ways of assessing validity
Face validity - simple way involves the extent to which items look like what a test claims to measure
Concurrent validity - assessed validity by correlating scores on a test with another test known to be valid
Predictive validity - assessed validity by prediction how well a test predicts future behaviour
Temporal validity - assessed to what degree research findings remain true over time
Scientific process
A means of acquiring knowledge based on observable measurable evidence
Scientific process method
Replicability - being able to repeat a study to check the validity of the results
Objectivity - observation made without bias
Falsification - that scientific statements are capable to be proven wrong
Paradigm
Consists of the basic assumptions, ways of thinking and methods of stuffy that are commonly accepted by members of a discipline or group
Paradigm shift + phases
Revolutionary changes in scientific assumptions
1) Inductive phase - observations yield info that is used to formulate theories as explanations
2) Deductive phase - predictions made from theories in the form of testable hypotheses, are tested and yield data that is analysed, leading to theory adjustments
What should be in a lab report
- Title page
- Abstract (last)
- Intro
- Method
- Results
- Discussion
- References
Title page
Indicate what the study is about, must include IV and DV
Shouldn’t be written as a question
Abstract
Write it last, provides concise and comprehensive summary of a research report
- Aim and rationale
- describe ppts and setting 5W
- describe method, what design, questionnaires or tests used
- Describe major findings statistics + significance
- contribution to knowledge (PA)
Introduction
Explain where hypothesis comes from
-general theory inc topic
-narrow down to specific and relevant theory
-logical progression of ideas, studies outlined lead to aims and hypothesis
AIM included in this para
HYPOTHESIS state the alternative hypothesis and make it clear inc variables
Method
Assume reader has no knowledge - so they could replicate it Past tense Don’t justify Subheadings -Design -Participants -Materials -Procedure
Results
Present the descriptive statistics followed by inferential statistics, don’t interpret results
Don’t include raw data
2dp
Significant or not
Discussion
Outline your findings Compare results to background material How confident can you be in the results, acknowledge limitations only if they can explain results obtained (not necessary) Constructive ways to improve Implications Idea for further research Concluding paragraph
Reference
Books:
Author, A. A. (Year). Title of work. Location: Publisher
Journal articles
Author, A. A., Author, B. B., Author, C. C. (Year). Article title. Journal Title, volume number(issue number), page numbers
Types of data
Quantitative - data occurring in numerical form
Qualitative - non numerical data expressing meanings, feelings and descriptions
Primary data
First hand research, done by the research, through data collected on his participants that’s never been published before
Secondary research
Data originally collected towards another research aim which has been published before
Meta analysis
A process in which a large number of studies which involve the same research within and methods of research are reviewed together and the combined data is tested by statistical techniques to assess the effect size
Content analysis
Method of quantifying qualitative data through the use of coding units
Adv and Dis of content analysis
Adv Ease of application Complements other methods Reliability Dis Descriptive Flawed results Lack of causality
Thematic analysis
A method of qualitative research that involves analysing data to identify patterns within it
Descriptive statistics
Provides summary of a set of data
Includes measures of central tendency and measures of dispersion
Measures of central tendency
Methods do estimating mod point scores in set of data
Median
Mean
Mode
Median and Adv/Dis
Central score in a list of ranked ordered scores Adv Not affected by extreme freak scores Easy to calculate Dis Not as sensitive as the mean Unrepresentative
Mean and Adv/Dis
Midpoint of the combined values of a set of data and calc by adding up all scores then dividing by total number of scores
Adv
Most sensitive measure of central tendency (accurate)
Uses all data
Dis
Less useful is some scores are skewed
Mean score my not be one of the actual scores
Mode and Adv/Dis
Most common number in a set of scores
Adv
Less prone to distortion by extreme values
Makes more sense then other measures eg average number of children is 2 not 2.5
Dis
Can be more then one mode
Doesn’t use all of the scores
Measures of dispersion
Measurement of the spread of scores within a set of data
The variability of scores
Range
Standard deviation
Range and Adv/Dis
Calculated by subtracting lowest value from highest value in set of data
Adv
Fairly easy and quick
Takes full account of extreme values
Can be distorted by exerted freak values
Does not show if data are clustered Or spread evenly
Standard deviation
Measure of the variability of a set of data, larger the SD the larger the spread of the scores
Calc
1)Add up all scores together and divide by number of scores to calculate mean
2)Subtract the mean from each individual score
3)Square each of these scores
4)Add all the squares scores together
5)Divide the sum of by the number of scores minus 1 (this is the variance)
6)Use a calculator to work put square root of the variance
Standard deviation Adv and Dis
Adv
More sensitive dispersion measure than the range since all scores calculated
Allows for the interpretation of individual scores m
Dis
More complicated to calculate
Less meaningful of data are not normally distributed
Percentages
Types do descriptive statistics that show the rate, number or amount of something within every 100, can be plotted on a pie chart
Bar charts
Shows data in for or categories to be compared like males and female scores conceding chocolate consumption
Categories X axis (hori)
Amount Y axis (vert)
Histogram
Similar to bar chart, main difference is histograms used for continuous data such as test scores
Continuous Scores - X axis (hori)
Frequency of these scores - Y axis (vert)
No space between bars
Pie chart
Used to show the freq of categories as percentages
Results/ data tables
Result: Summarise the main findings
Often uses Measure of dispersion and central tendency
Data tables presents raw unprocessed scores
Normal distribution
Data with an evenly distribution of scores either side of the mean
Skewed distribution
Data that does not have an even distribution of scores either side of the mean
How to check if it’s normally distributed
Examine visually - look at the data to see if most scores are clustered around the mean
Calculate measures to central tendency - calc mean mode and median to see if similar
Plot the frequency distribution - plot the data on a histogram to see if it forms a bell shaped curve
Causes of skewed distribution
Outliers (extreme freak scores)
Positive skewed : occurs when there is a high extreme score/s
Negative skewed : occurring when there is a low extreme score/s
Sign test
A non parametric statistical test used for experiments where data is at least nominal and a repeated measures design has been used
Correlation coefficient
Correlation study produces it
‘A numerical value showing the degree to which two co variables are related
Range from +1 perfect positive correlation to -1 perfect negative correlation
Little correlation will be near to zero
Data from correlation analysed by spearmans rho and pearsons
Inferential testing
Statistical procedures that makes predictions about populations from mathematical analysis of data taken from a sample
How to decide what inferential test needs to be used
1) whether a difference or relationship between 2 sets of data is being tested for
2) what level of measurement the data is (nominal, ordinal and interval/ratio
3) what design has been used : IGD, RMD(inc MPD)
Levels of measurement
NOIR
Nominal data
Ordinal data
Interval/ratio
Nominal
Involves counting frequency data eg how many days of week were rainy or sunny
Tally charts used to record data
Nominal data crudest, most uninformative eg does tell us how many hours sunny
Ordinal
Involves ranking data into place order with rating scales often being used to achieve this
More informative than nominal but lacks being fully informative eg athletes finishing places in race 1st, 2nd … but doesn’t Inform the distances between individuals
Similarly one persons subjective rating of 7 may be very different to another rating of 7
Interval / Ratio
Standardised measurement units like time, weight, temp abs distances are interval/ratio measurements
Most informative and accurate as they use equal measurement intervals
Probability
The likelihood is events being determined by chance
Type | and type || errors
|| when a difference / relationship in a data set is rejected but actually does exist
Eg significance levels been set too low
Null hypothesis wrongly rejected
when a difference / relationship in a data set is accepted as a real one but in fact is not
Eg significance level has been set too high so null hypothesis is wrongly rejected
Sign test
Used when a difference is predicted between two sets of data, the data is of at least nominal nature and RMD has been used
Chi squared
Used when a difference is predicted between two sets of data, data is at least of nominal and IGD has been used, Chi squared as a test of association (relationship)
Mann whitney
Used when a difference is predicted between two sets of data, data at least ordinal level and IGD has been used
Wilcoxon signed matched ranks
Used when a difference is predicted to occur between two sets of data, data at least ordinal level and RMD or MPD has been used
Independent (unrelated) t-test
Used when a difference is predicted between two sets of data, data is normally distributed data is of interval/ ratio level and IGD has been used
Repeated (related) t-test
Used when a difference is predicted between two sets of data, data is normally distributed data is of interval/ ratio level and A RMD or MPD has been used
Spearmans rho
Used when a relationship (correlation) is predicted between two sets of data, the data is of at least ordinal level and the data are pairs of scores from the same person or event
Pearson product moment
Used when a relationship (correlation) is predicted between two sets of data, the data is of at least interval/ratio level and the data are pairs of scores from the same person or event
One tailed or two tailed
A one tailed test is where you are only interested in one direction. Of a mean is x, you might want to know if set of results is more or less than x. A one tailed is more powerful than a two tailed test as you aren’t considering an effect in opposite direction
One- directional
Two - non directional
Tests that need observed value to be equal or less than the critical value to be accepted, allowing null hypothesis to be rejected
Mann whitney
Wilcoxon
Sign test
Tests that need observed value to be equal or greater than the critical value to be accepted, allowing null hypothesis to be rejected
Chi squared Independent (unrelated) t-test Repeated (related) t-test Spearmans rho Pearson product moment
How to remember inferential tests
I Read Comics
Naughty Children Swear Continuously
Only Men Want Sausages
Reading Usually Raises Psychology marks