Research Methods Flashcards
Aim
What the research wants to investigate.
Hypothesis
Predictions about what the research wants to investigate.
Independent Variable
Can be manipulated.
Dependent Variable
Can be measured.
Operationalising Variables
To make the variable measurable.
Experimental hypothesis
The prediction of what you expect to find.
Null hypothesis
It predicts that there will be no difference or no significant difference between the 2 groups of the study.
Alternative hypothesis
Testable prediction of what you expect to find.
Two-tailed or non-directional
It’s used when the result cannot be certain.
One-tailed or directional
When previous research suggests that certain results will happen.
Pilot studies
It’s a small scale trial run of a specific research investigation to test out the planned procedures and identify any flaws and areas for improvement, before time and money are invested in carrying out the main study. It’s carried out on a small no. of ppts to find out whether there are any problems with:
- the design;
- clear instruments;
- measuring instruments;
- direct experiencing and feedback;
Bickman’s Study- Evaluation
Weakness: lacks ethical guidelines
Strength: easy to control and replicate as there’s a standardised procedure.
Independent Measures- Evaluation
Strength: number of ppts needed
Weakness: ppts Variables
Repeated Measures - Evaluation
Strength: High validity
Weakness: Demand characteristics
Matched pairs - Evaluation
Strength: Reliable
Weakness: Difficult to match ppts
Ethical Issues
They arise when there’s a conflict between the needs of the research/ researcher and the ethical rights/ dignity of the ppts.
Informed Consent
It’s when ppts make an informed decision as to whether they wish to participate in the study.
Debrief
Informing ppts after study and give them the right to withdraw.
Deception
It’s similar to lack of informed consent but is more general issue referring to any instance in which the ppts have been lied or misled.
Protection from harm
Ppts should not be put in a situation that could cause them to be more at risk of physical/ psychological harm than in their day to day lives.
Confidentiality
Results must not be traceable back to the ppts.
Privacy
They should only be studied if they are in a place in which they expect to be seen.
Right to withdraw
Ppts should have a way of contacting back the researcher to make a different decision on their data.
Zimbardo’s Study - Ethical Issues
Right to withdraw -> prisioneiros breakdown;
Privacy -> naked -> they were arrested in front of everybody else;
Informed consent -> don’t explain study;
P.F.H. -> Mental breakdown + physical abuse;
Conducting a research
BPS has a code of ethics and they provide guidance on how to plan and conduct research in an ethically acceptable way. It looks for 4 main themes:
- Respect -> show respect for ppts;
- Integrity -> Demonstrate scientific integrity = highly quality;
- Social Responsibility -> Improve understanding of human nature;
- Maximise Benefit & Minimise Harm -> harm shouldn’t be greater than the harm faced in real life;
Random Sampling
Everyone has an equal chance of being chosen. Individuals are given a number and a computer chooses numbers at random.
Advantage: highly representative
Disadvantage: time-scale of gathering info may be too long.
Opportunity Sampling
A technique that involves just giving your questions to anyone who happens to be available and chose to take part.
Advantage: inexpensive way of ensuring sufficient no. of a study.
Disadvantage: highly unrepresentative
Stratified Sample
Population is divided in subgroups according to certain characteristics. Then randomly draw a sample from each subgroup in proportion to the amount of people in the population.
Advantage: specific groups are represented in the sample
Disadvantage: complex and effort
Systematic Sampling
Use of Sampling frame but younger every ’nth’ person e.g: every 10th person on registry
Advantage: can be generalised
Disadvantage: not representative
Self selected sampling
Ppts selected on the basis of the ppts’ own action at arriving at the sampling point. It can either be volunteering or people in a certain placed being asked to do it.
Advantage: people are likely to be willing to take part
Disadvantage: unrepresentative of population
Target population
The part of the population from which the sample is selected.
E.g: single mothers with 1 child that live in Birmingham
Field experiment
When the researcher still manipulated the IV to compare the DV but it’s done in a natural setting (town centre, supermarket etc.) and it’s a standardised procedure.
Strength: Conducted in a natural setting so people will act in a natural way.
Weakness: Can’t control for outside factors - extraneous Variables
Lab experiment
Takes place in a controlled setting. Researcher manipulated the IV and measuring the effects of that change on the DV. Standardised procedure is used.
Strength: Control situation + be sure that the DV’s a result of the IV’s.
Weakness: People might not behave in a natural way.
Natural Experiment
Naturally occurring IV is used, so the researcher can’t manipulate. Eg: the researcher can’t manipulate how long a child attends nursery because it’s decided by someone else.
Strength: IV is naturally occurring - so gives high levels of validity and reliability
Weakness: More difficult to recruit ppts
Non- participant observation
A type of observational study whereby the researcher does not join in with the activity being observed.
Strength: Natural behaviour
Weakness: Time consuming/ No control
Participant observation
A type of observation study where the observer is also a ppt in the activity being studied.
Strength: No waiting
Weakness: Biased/ Different behaviour
Covert observation
The ppt’s inward that an observation is taking place.
Strength: Behaviour might be more realistic
Weakness: Unethical
Overt observation
The ppt is aware they are being observed.
Strength: Ethical
Weakness: Change in behaviour
Event Sampling
Recording the number of instances of a behaviour. Need to create a coding scheme.
Coding System
A structure observation is where the researchers design a type of coding scheme to record the ppts’ behaviour. Provides quantitative data. Coding schemes are ways of categorising behaviour so that you can code what you observe.
Time Sampling
Observing behaviour at set time intervals and recording it.
Questionnaires
Type of self report method. They consist of a series of pre-set questions that are often written down. This means the questions are the same for everyone who takes part in te questionnaire. It’s a non-experimental method.
Writing Questions - Rules to follow or things to avoid
Keep it as short as possible. If people get bored - they will put anything - lower validity.
- Start with easy questions.
- Avoid jargon e.g: ‘novels’ - do a pilot study if unsure.
- Avoid leading questions - can lead to ‘social desirability bias’
Open questions - Questionnaire
This are questions which require the participant to write something for their answer. This means that no 2 responses will be the same. They produce qualitative data.
Closed Questions - Questionnaire
This are the questions that can be only answered in a limited number of ways:
- Tick box/ Circle questions
- Rate on a scale of 1 - 5
Three produce qualitative data.
Interviews
Gathering info in response to face-to-face questioning
Structured ->interview
Set format of standardised questions
Strength: Easy to replicate, Quick to conduct
Weakness: Aren’t flexible, Lack detail
Semi- structured -> Interview
Set questions but these can be varied and expanded upon
Unstructured -> Interview
No set format the questions allows interviewer to follow the natural progression of the conversation
Strength: more flexible, qualitative data, increased validity
Weakness: time consuming, employing and training are expensive
Case studies
An in-depth study that gathers a lot of detail about 1 person or a small group.
Strength: Qualitative data, Validity, Ecological Validity, Triangulation
Weakness: Subjectivity, Generalisation, Lack of replication, Time-consuming
Triangulation
Gathering data using a variety of sources.
Correlation
It’s a measure of how strongly 2 or more variables are related to each other.
There’s no IV or DV in correlation, just co-occurring Variables.
Correlation coefficient
A number between -1 and 1 that tells us how strong the relationship is. \+0.1 perfect positive \+0.8 strong \+0.5 moderate \+0.3 weak 0 no correlation -0.3 weak negative correlation -0.5 moderate -0.8 strong -1.0 perfect
Internal Reliability
To the extent to which a measure is consistent within itself
Eg. If 1 part of the study shows that you’re introvert, another part should show the same.
*Factors affecting:
- observations;
- self reports;
- correlations;
External reliability
Produces same results each time the test is carried out.
- Population Validity
- Ecological Validity
- Temporal Validity
Reliability
When something can be repeated to get the same results it’s considerable reliable. Whether or not the measured used is suitable.
Inter-rated reliability
To consider psychological research. If there’s high inter-rate reliability this essentially means that 2 or more individuals have a high agreement on a score and therefore the measurement of behaviour is reliable.
Split-Half Method
Applies mainly to self-reports.
It means that results in 2 halves of a questionnaire are similar; the test is reliable
Demand characteristics
Please you, screw you effect
Investigator effects
Age, voice tone, gender
Situational variable
Lightening, weather, room conditions
Content Analysis
The method by which you take qualitative data and analyse it in order to draw conclusions. This involves converting qualitative data into quantitative data.
Process of content analysis
Step 1: Sampling
Step 2: Coding Units
Step 3: Pilot Study
Step 4: coders
Process of content analysis- Evaluation
Strength: Reliability - coding system and training difference researchers allows them to compara data
Weakness: Reductionism - converting qualitative data to numbers is reductionist
Weakness: Illusion of objectivity - transforming qualitative data into numbers appears to be scientifically but it may be biased
Peer Review
Very important throughout the whole of the scientific community and it has 2 major functions.
- Researchers get to read other people’s studies & keep in tou h with new ways of thinking and scientific developments.
- Studies submitted for publication are subjected to critical appraisal,, which acts as a break to ensure poor quality research doesn’t enter the public domain.
Quality check - peer review
Check the report is in the correct format an that the written clarity is to a high standard.
Check method - peer review
Ensure that the method is reported in such a way to allow replication of the research at a later date.
Verify conclusions - peer review
Ensure that the conclusions reported are appropriate given the methodology and the findings.
Fraud research - peer review
Plagiarism
Falsification
Fabrication
Plagiarism
The act of stealing someone else’s work and attempting to pass it off as your own.
Falsification
Making date false by altering or adding to.
Fabrication
Making up results or data and recording or reporting them.
Peer review - Evaluation
- Strength: Quality check- the info can be trusted
- Strength: Benefits- careers of research
- Weakness: Slow, expensive, subjective
- Weakness: An expert might be hard to find
Product bias
Variety of forms
Eg: gender bias, institution bias or the reviewer may have a different viewpoint from the research.
File drawer phenomenon
This is where the peer review tends to favour positive and as such those pieces of research where the null hypothesis has been supported, have been found to be rejected or ‘filed in the draw’. This leads to an unrepresentative view of the findings.
Objectivity
Although al psychologist try to objective, a reviewer nah favour an opposing view making them less likely to provide an unbiased view.
Preserving the status quo
It can be argued that science is conservative and resistant to large changes and so peer review could act to maintain the status quo, as such results that don’t fit in with current knowledge may be rejected.
Measures of dispension
Measures of ‘spread’.
Looks at how ‘spread’ out the data is.
Range
Difference between highest and lowest numbers.
Standard deviation
Better measure of ‘spread’ thanthe range because it’s less affected by extreme values.
However it does take longer to calculate
Enables us to compare standard deviation of 2 or more groups to help draw conclusions.
|——————————
S= |£ (x- x^)2
———————————
N
Square root the Sum of (each value in data set minus mean of all values) all divided by the number o values
Skewed Distribution
There are called skewed distribution when it’s not symmetrical at the mean or median or mode. A slew can be positive or negative.
Types of skews
Negative skewed: left foot shape
- mean, media, mode (increasing shape)
Normal skewed: no skew normal curve shape
Positive skewed: right foot shape
-mean, median, mode (decreasing shape)
The skew depends on whether the mode falls higher or lower than the median.
Scatter graph
Used with correlations where the relations of 2 variables is summarised.
Bar chart
Used to represent data from frequency tables, mean scores or total levels. kept separate from each other.
Histogram
Used with interval or ratio data. No gaps; it’s a continuous data set.
Tally graph/ Frequency Table
Used to count things. Used in observations and content analysis.
Line graphs
Used as an alternative to histograms. These are used to show the results from 2 or more conditions at the same time.
Pie chart
Used when we have percentages. Each segment represented a percentage of the total.
Inferential Statistic tests
- Sign Test
- Chi Square
- Mann Whitney
- Wilcoxon
- Spearman’s Rho
- Pearson’s R
- Related t- test
- Unrelated t-test
Table of Inferential stats test
Data | Related Design | Unrelated Design | Correlation
——————————————————————————-
Nominal | Simon | Cowell | Considerately
| Sign Test | Chi square | Chi square
——————————————————————————————
Ordinal | Wants | More | Singers
| Wilcoxon | Mann. Whitney | Spearman’s Rho
——————————————————————————————-
Interval | Receiving | Unbelievable | Praise
| Related | Unrelated | Pearson’s R
——————————————————————————————
Related design
Repeated measures and matched pairs
Unrelated design
Independent groups
Nominal
Can be put in categories
Eg. Small/tall/medium
Ordinal
Can be put in order
Eg. Grades (highest to lowest)
Interval
Can be measured and compared the exact time
Eg. Ruler
Replicability
The extent to which a study can be repeated to find the same or similar results.
Ensures validity, reliability and objectivity.
Objectivity
Free from bias.
Importance as it highlights cause and effect.
Robust enough to be scientific
Theory Construction
Induction - through observations and testing conclusions are drawn and interpreted into a general theory.
Deduction - general theory forms observations which are tested to make a specific theory.
This allows us to create hypothesis which we can test deduction considered more scientific as based on facts.
Empirical method
Gaining knowledge through experience.
Eg. Observations and experiments to gather facts as evidence.
Paradigms and Paradigm Shifts
It’s a set of principles, methods and techniques which define a scientific discipline.
The subject must have some basis assumptions and principles (a paradigm)
External Validity- Ecological
Can you generalise results to the real world.
External validity- Population
Can you generalise findings to the wider population.
External Validity- Temporal
To ability to generalise your results to different time periods.
Face Validity
Judgement about whether a test seems to be valid at first glance.
Concurrent Validity
A test shows concurrent Validity if it shows similar findings to another existing measure.
Concurrent
If the score correlate as a strong positive correlation (+0.80) then themed valid.
Assessing reliability - Test retest
Giving the same test to the same people in the same condition but on separate occasions. It must be enough time for them to not be able to complete it with memory, but not too long that they have changed as a person.
Must have a +0.80 correlation to be reliable.
Assessing reliability- Inter-observer
Different researcher may interpret a situation differently (subjectivity bias), so this must be established so more than 1 researcher interpret a situation in the same way. +0.80 and significant to show reliability.
Content - analysis
- Data is collected;
- Researcher reads through or examines the data making themselves familiar with it;
- The researcher identifies coding units;
- A tally is made of the no. of times a code unit appears.
- Check the reliability of the consent analysis by correlating one researcher’s scores with another.
Thematic analysis
- Collecting data & transcribing it;
- Researcher familiarises themselves with it;
- The researcher will then code the data and put into categories;
- The researcher will then write up the final report, using quotes from the data to illustrate the themes.
Content & Thematic analysis -Evaluation
Strength: can avoid ethical issues
Strength: high ecological validity
Weakness: people are studied indirectly
Weakness: bias reduces objectivity
The Sign Test - Conditions
Conditions for use:
- Related Design = repeated measures/ matched pairs
- Nominal data
- Test of difference
The Sign Test- Calculating
Step 1: check the study meets all the conditions for use
Step 2: identify whether hypothesis is 1-tailed or 2-tailed
Step 3: calculate the difference between the conditions
Step 4: put the ‘sign’ into the sign column (+-=)
Step 5: count to the no. of each sign and ignore any equal signs/ no difference
Step 6: use the lower no. of signs as your test statistic/ observed/ calculated value ( S value)
Step 7: use table to find critical value
Step 8: interpret results
Step 9: Accept or reject hypothesis and write up results
Write a psychological report
Title Abstract Introduction Method Results Discussion Reference
Referencing a book
Author’s surname, initial (date published), title of book, place of
—————
publisher, publisher.
Referencing of journal
Author’s surname, initial (date published), title of article , journal
———-
title, volume (issue number, page numbers used).
——
