Research Methods Flashcards
What is ecological validity?
If it reflects real life
Define Operationalisation
When variables are clearly defined in terms of how they can be tested and measured (should be in the hypothesis) e.g reaction time in secs, number of…
What is an experimental method?
The manipulation of variables to establish cause and effect relationships
Define a Directional hypothesis
States the direction of the predicted difference (one-tailed)
Define a Non-directional hypothesis
States that there will be a difference but we don’t know what direction the difference will be. (two-tailed)
Define a Null hypothesis
Statements of equality, statements of no relationship, statements of no casual relationship.
What is an extraneous variable?
Any variable (other than IV) that may affect the DV if not controlled.
Define investigator effect
Any investigator behaviour that may effect the outcome.
What are demand characteristics?
Any cues from the researcher or situation that might reveal the purpose to the participants, therefore may act in a different way.
Define standardisation
Using the same formalised procedures and instructions for all participants
What is randomisation?
Using ‘chance methods’ to control investigator effects in the design of research
Name the two types of extraneous variables
- Participant (individual differences)
- Situational (experimental situation)
may effect the DV
Define the experimental design ‘matched pairs’ and its + and -
Pairs of p’s are matched on variables that may effect the DV (e.g IQ or age), then assigned to condition A or B.
+Order effect is not an issue
-Time consuming and expensive to match, especially id a pre-test is needed
Define the experimental design ‘independent groups’ and its + and -
Two separate groups of p’s experience two different conditions.
+Order effect is not an issue
-Twice as many p’s are needed, more time and money needed
Define the experimental design ‘repeated measures’ and its + and -
All p’s will experience all conditions of the experiment.
→Counterbalancing: Half the p’s experience one order, the other half in the opposite order
+Less people needed
-Order effect may occur
What is Validity?
Whether something is true or legitimate
Explain the 2 types of validity and the 3 subsections if these
Internal- within the research setting
External- outside the context of the study
-ecological: places/settings
-population: people
-temporal: time
What is reliability?
A measure of whether something stays the same.
Can do ‘test retest reliability’ where the study is repeated
Define a Laboratory experiment and evaluate
Highly controlled environment, doesn’t have to be a lab.
+High control over EVs
+Replication more possible
-Tasks lack realism, aware being tested (demand characteristics)
Define a Field experiment and evaluate
The IV is manipulated in a natural environment.
+Realistic environment, high external validity
-Less control over EV
-No consent
-Replication is hard
Define a Natural experiment and evaluate
Like a field, yet there is no control over the IV
+High external validly
+Can apply to real life
-No control over extraneous variables
Explain a Quasi experiment and evaluate
The IV is based on existing differences of the people e.g age or gender. The DV may be naturally occurring or developed.
+Controlled
-As the IV is not in control of researcher, cannot claim it caused change
Define what a Confounding variable is
A type of extraneous variable that vary systematically (relates) with the IV, so we can’t be sure of the true source of change to the DV
What is random allocation with regards to Independent groups?
Ensures all participants have the same chance of being allocated to a condition
BPS produced a code of ethics, what does it stand for?
British Psychological Society
What are the ethical considerations, and can you explain them?
RIP DDC
Right to withdraw -allowed to leave at any point
Informed consent -consent to take part, fully aware of aims and consequences
Protection from harm -from psychological and physiological harm including distress and embarrassment
Deception -not be deceptive, can’t lie about aim
Debrief -to return them to their original state, reveal details
Confidentiality -their info is protected
What are the sampling methods?
VOSSR
Volunteer (self selection)
Opportunity
Systematic
Stratified
Random
What is volunteer sampling and evaluate it?
Participants select themselves
+Easy to conduct
-Unrepresentative sample
What is opportunity sampling and evaluate it?
Selecting anyone available to take part
+Quick and easy
-May not be representative of target population
What is systematic sampling and evaluate it?
When every nth person is selected from a list
+No bias
-Time consuming
What is stratified sampling and evaluate it?
Same percentage of people in a subgroup in the sample as the target population
+Good cross section of people, reflects population
-Time consuming
What is random sampling and evaluate it?
A list of the target population, into a random name generator
+No bias
-Unrepresentative
Define Target population
A group of people who are the focus of a researchers interest
Define Sample
A group who take part in the research taken from the target population
What is a Pilot study?
A small scale version of an investigation that takes place before the real investigation is conducted.
Why might you want to do a Pilot Study?
-Identifying extraneous variables
-Are you actually testing what you want to test
-Do materials work
-Is it a replicable procedure
-Do you need to make any changes
What is a single blind procedure?
Ps are not aware of the aim or what condition they are in if there is on
+No demand characteristics
What is a double blind procedure?
Neither the Ps or the researcher are aware of the true aim of the research
+No investigator effect
-No bias
Define what control groups are
A baseline for comparison- if there is a significant difference between them and the experimental group, we know it was the IV.
Define self-report methods
Research methods that ask Ps about their own thoughts, feelings and opinions.
-Interviews
-Questionnaires
What are closed and open questions?
Closed- Yes or no answers, short
Open- Longer answers and more detail
There are 3 types of closed questions, what are they?
- Likert scale
- Rating scale
- Fixed choice option
What is a Likert Scale?
Respondent indicates their agreement with a statement based on a scale of points
Strongly agree
Strongly disagree
What is a rating scale?
Like a likert scale but respondents identify a value that represents their feelings
- 5
What is a fixed-choice option?
A list of possible options and respondents indicate those that apply to them.
-Horror
-Rom com
-Comedy
What are some considerations when designing questions?
-No leading questions
-No bias
-No double barrelled questions
-No harm or distress
-Understandable
Define what a Correlation is
A researcher investigates an association between two co-variables
What are co-variables?
The variables investigated in a correlation
What is a positive correlation?
As one variable increases the other increases
What is a negative correlation?
As one variable increases, the other decreases
What does no correlation mean?
No relationship between variables
Evaluate correlations
+Practical and less time consuming
+Open doors for new research
-Doesn’t establish cause and effect
-More co-variables may cause effect
What is Qualitative data?
Information expressed in words not numbers
-Open questions
-Interviews
What is Quantitative data?
Information that can be counted
-Closed questions
-Multiple choice questions
Evaluate Qualitative data
+More detail
+High external validity
-Subjective
-Difficult to analyse = time consuming
Evaluate Quantitative data
+Easier to analyse
+Objective
+ Less chance of bias
-Lacks detail
-Less chance of real world application
What is primary data?
Information that has been obtained first hand by the researcher
What is secondary data?
Information from someone other than the researcher
Evaluate the use of primary data
+Suits the aim of the research
+Authentic
-Takes more time and effort
-Possibly costly
Evaluate secondary data
+Already checked, less time and money
+Easy to access
-May not exactly for what the researchers want
-May be out of date or poor quality
What is a Meta-analysis?
Combining the findings of a number of studies with the same aim/hypothesis. The results are pooled and a joint conclusion is produced.
Evaluate the use of a meta-analysis
+Larger, more varied sample
+Generalisable, increases validity
-Open to publication bias- may leave out unwanted negative data.
What are the observation types?
-Naturalistic
-Controlled
-Covert
-Overt
-Participant
-Non-participant
What is a naturalistic observation, + and -?
In a setting where it would normally take place.
+High external validity, done in a natural environment
-Extraneous variables
What is a controlled observation, + and -?
Structured environment e.g lab setting
+More control over extraneous variables
-Lacks realism
-Unnatural behaviour
What is an overt observation, + and -
Knowing they are being watched.
+Ethically acceptable
-Demand characteristics
-Unnatural behaviour.
What is a covert observation, + and -
Unaware behaviour is being watched and recorded.
+Natural behaviour
-Ethical issues
What is a participant observation, + and -?
Researcher who is observing is part of the group being observed.
+More insightful, increases validity
-Researcher may lose objectivity as they may identify too strongly with the participants.
What is a non-participant observation, + and -?
Researcher observes from a distance, not part of the group.
+Researcher can be more objective and less likely to identify with Ps.
-May lose valuable insight
What is observer bias?
An observers reports are by biased by what they want to see.
What is inter-observer reliability?
Many researchers conducting the observational study to test its reliability.
Explain the ‘Unstructured’ observational design, + and -?
Consists of continuous recording where the researcher writes everything they see during the observation.
+More richness and depth of detail
-More difficult to analyse and record
Explain the ‘Structured’ observational design, + and -?
Researcher quantifies what they are observing, using predetermined list of behaviours and sampling methods (behavioural categories)
+Easier to analyse the data
-Not much depth and detail
What are behavioural categories?
The target behaviour observed is broken up into more precise components which are observable and measurable (operationalisation)
What is ‘Time-sampling’ in structured observations, + and -?
Recording of behaviour within a time frame that is pre-established.
+Reduces number of observations, less time consuming
-Small amount of data, normally unrepresentative of observation.
What is ‘Event-sampling’ in structured observations, + and -?
Counting the number of times a particular behaviour is carried out.
+Good for infrequent behaviour
-Important details of complex behaviour may be overlooked
What is an observation?
Non-experimental technique, where the researcher watches and records behaviour without manipulating the IV
What are the 3 measures of central tendency in descriptive statistics?
The averages:
Mean
Median
Mode
What are the 2 measures of dispersion in descriptive statistics?
Spread and Variation:
Range
Standard deviation
Evaluate the mean
+ Includes all values, so more representation
- Easily distorted by extreme values
Evaluate the median
+ Not effected by extreme scores
- Ignores extreme values that may be important
Evaluate the mode
+ Easy to calculate
- Not representative of all the data
Evaluate the range
+ Easy to calculate
- Only takes into account the 2 most extreme values, so unrepresentative
How do you calculate the mean, median, mode, and range?
Mean: Add up all values, divide by total number of values
Median: The middle value
Mode: The value that occurs the most
Range: The largest subtract the smallest value
What is ‘Standard deviation’?
Tells us how far scores deviate from the mean.
Low SD: data points are clustered around the mean
Large SD: data points are widely spread around the mean
How can you display quantitative data?
-Histogram
-Pie chart
-Line graph
-Bar chart
-Table
-Scatter graph
In a bar chart there are 3 distributions, what are they?
Normal
Positive skewed
Negative skewed
Explain the normal distribution on a bar chart
-Mean, median and mode are all at the midpoint of the curve
-Data is distributed symmetrically
-Scores dispersed consistently around the midpoint
-Express in standard deviation
Explain a positive skewed distribution of a bar chart
-Curve no longer bell shaped
-Scores are dispersed consistently around lower scores and not the mean
-Mean, median and mode are no longer the same
-Extreme high scores are having an effect
Explain negative skewed distribution in bar charts
-Curve not bell shaped
-Scores are dispersed consistently around higher scores not the mean
-Mean, median and more not the same
-Extreme low scores are having an effect
Describe tables as a presentation of data
Used to summarise data so whoever is reading it can make sense of the numbers as they have been converted into descriptive statistics
Describe scatter graphs as a presentation of data
-Measures relationship between 2 variables one on
-An ‘X’ where 2 variables meet
-Reveals correlation or pattern
Describe graphs as presentation of data
-Summarise quantitive data
-Easily see patterns
Describe bar charts as a presentation of data
-Summarise data that is already categorised
-Mean or frequency on y-axis
Describe a histogram as a presentation of data
-Represents data on a continuous scale
-Columns touch
-Height shows frequency
Describe the process of a sign test
-Do the same task twice but change something the 2nd time round
-Work out the sign difference better or worse (+ or -)
- S is the less frequent sign value
- N is the number of participants but ignore = values
- Significance level generally is p<0.05
- Use a table of critical values and is significant is the critical value is lower than S
Significant= accept experimental hypothesis
Not significant= accept null hypothesis
What conditions do we need to use a sign test?
-Looking for a difference
-A repeated measure design
-Nominal data (categories)
What is an estimate?
Used to reduce ball park figures, to create a broad idea of what the results would be.
What are ‘order of magnitude calculations’?
Another kind of estimate using standard form and comparing exponents.
What is peer review?
The assessment of work by others who are specialists in the same field to ensure that any research intended for publication is of high quality
What are the aims of peer review?
-To validate quality and relevance of research: hypothesis, methodology, statistical tests, and conclusions
-To allocate research funding:
decide if funding should be awarded.
-To suggest amendments or improvements: suggest minor revisions, or conclude the work is inappropriate for publication and should be withdrawn.
Evaluate the use of peer review
+ Anonymity: more honest review, but can use to criticise their competition
-Publication bias: prefer to publish positive results to increase credibility, a false impression
-Burying groundbreaking research: may not want to publish anything that opposes their own research.
What are some implications of research?
Attachment: Bowlby- secure attachment with mothers, but other research suggests father attachment is just as important , so more mums at work increases the economy.
Mental Health: more awareness of disorders, more people off work with it, costs a lot, decreases the economy.
What is a case study?
Detailed study into the life of a person which covers great detail into their background.
Give an example of a case study
HM- procedural memory was intact but episodic wasn’t
Strengths and weaknesses of case studies
+Detail
+Form basis for other research
-Not generalisable
-Time consuming
What is content analysis?
Studying human behaviour indirectly by studying things that we produce e.g TV adverts or newspapers.
How might you conduct content analysis?
-Hypothesis
-Coding system
-Record data
-Use thematic analysis
presence of certain words, themes or concepts within qualitive data
Evaluate content analysis
+Strong external validity
+Easy replication
-Observer bias
-Interpretive bias
What are the levels of measurement?
-Nominal data
-Ordinal data
-Interval data
What is nominal data?
In the form of categories, discrete
What is ordinal data?
Represented in ranking form
What is interval data?
Based on numerical scales which include equal units of precisely defined size.
What is included in a scientific report?
Abstract: key details of research report.
Intro: info on past research and relevant theories are mentioned.
Method: description of what researchers did
Results: findings
Discussion: considers findings and what they mean
Referencing: list sources and quotes
How would you reference books and journals?
Books:
author, date, title of book (italics or underlined), place of publication, publisher.
Journals:
author, date, title of article, journal name, journal volume, issue number, page range
How would you choose a statistical test?
Design: unrelated (independent) or related (matched pairs, repeated masures)
Level of data: ordinal, nominal, or interval
Difference or correlation?
How do you remember the stats tests table and what is it?
Carrots Should Come
Mashed With Swede
Under Roast Potatoes
Top headings: Unrelated, related (test of differences), test of correlation
Side headings: Nominal, Ordinal, Interval (No, I)
Chi-Squared, Sign test, Chi-Squared
Mann-Whitney, Wilcoxon, Spearman’s Rho
Unrelated t-test, Related t-test, Pearson’s rank
What are the use of stats tests?
Used to determine whether a significant difference or correlation is discovered using the calculated value- tests hypothesis and significance
What is significance?
Lets us know how sure we are about a correlation or difference existing
What is probability?
How likely it is for an event to happen
What is a type 1 error?
Optimistic-
the incorrect rejection of a null hypothesis which is actually true. Claim to have found a difference but there isnt one (false positive)
What is a type 2 error?
Pessimistic-
The failure to reject the null hypothesis, claim there is no sig difference when there is (false negative)
What are the key points to features of science?
-Paradigms and paradigm shifts
-Theory construction and hypothesis testing
-Falsifiability
-Replicability
-Objectivity and the empirical method
What is a paradigm and a paradigm shift?
Paradigm= set of shared ideas and assumptions within a scientific discipline.
Paradigm shift= significant change in these central assumptions within a scientific discipline, resulting in a scientific resolution.
What is a theory?
Refers to a set of general principles and laws which can be used to explain specific events or behaviours
Explain theory construction
Gathering evidence from direct observation during investigations
What is deduction? give an example
Deriving new hypotheses from an already existing theory
e.g. Baddeley and Hitch modified the working memory model and added episodic buffer
What is falsifiability? give an example
Principle that states a theory cannot be considered scientific unless it allows itself to be proven untrue
e.g. Freud’s concepts, like Oedipus’s complex
What is replicability?
The extent to which scientific methods and their results can be repeated by other researchers across other contexts and get the same results
What is objectivity?
When all possible biases from the researcher are minimised so that they don’t influence the research process.
What is the empirical method?
When evidence is collected through making direct observations and through direct experiences
Evaluate features of science
+ Produces intuitive results which are against common sense
+Give scientific credibility
-Experiment interpretations can be subjective
-Not all research is generalisable e.g. case studies
What is concurrent validity?
Occurs if a test is similar to an older test that already has well established validity
What is face validity?
If a test appears to be valid at first appearance in spite of how well it works in a real world secenario.
it appears to measure what its supposed to measure