PS125 Research and Statistical Methods Flashcards
Stages of the scientific method in psychology
Stages of the scientific method in psychology = Question, Hypothesis, Prediction, Experiment, Analysis, Interpretation
What are the key concepts of the scientific method in psychology?
Key concepts of the scientific method in psychology:
Flasifiability
Null Hypothesis
“Testing to destruction”
Reliability
Validity
Testing reliability in psychological research
Testing reliability in psychological research:
Test-Retest
Inter-Rater
Internal Consistency
Parallel Forms
Types of validity in psychological research
Types of validity in psychological research:
Face
Construct
Concurrent
Predictive
What are the principles of scientific research?
Principles of scientific research:
Beyond observation and intuition
Generalisation
Testable, falsifiable hypotheses
Replicability
What are the main principles in ethical research as outlined by the BPS?
Ethical Research:
Informed Consent
Privacy
Confidentiality
Empathy
Competence
Responsibility
Integrity
Quantitative methods in psychological research
Quantitative methods in psychological research:
Numerical Data
Questionnaires + Scales
Experimental Method
Qualitative Methods in psychological research
Qualitative Methods in psychological research:
Lived Experience
Detailed Observation
In-depth Interviews
Ways of which scientific research is communicated
Ways of which scientific research is communicated:
Research Reports
Journal articles
Peer Review
Statistical Method definition
Statistical Methods = the collection, presentation, analysis and interpretation of data
Inferential statistics in psychological scientific research
Inferential statistics = makes generalisations about a population by studying a sample
-Can draw inferences about parameters from statistics
What are parameters?
Parameter = descriptive measure of a population
What are statistics?
Statistic = descriptive measure of a sample
A good hypothesis should…
A good hypothesis should:
-be stated in a declarative form
-place a relationship between variables
-reflect a theory in which they are based
-be brief
-be testable
What are the different research hypotheses?
Research Hypotheses:
-State a difference or inequality in relationships
-State a difference or inequality across groups
Differentiate non-directional and directional hypotheses
-Non-directional = no previous research, two-tailed tests
-Directional = previous research, one-tailed tests
Define a null hypothesis
Null hypothesis = states there are no relationship or difference between variables
-> Variables are equal or unrelated
-> Precise and well defined
-> Can be compared to to determine if its due to chance or not
Null hypothesis testing
Null hypothesis testing;
-> Either rejected or accepted
-> Accepting a null hypothesis can be done via confidence intervals or bayesian hypothesis tests
What is probably in scientific psychological research?
Probability = the degree of confidence we havein stating that a particular outcome may not have occured due to chance alone
Measured between 0 and 1
Calculating probability
Probability corresponds to relative frequency
Relative frequency = frequency / N
High probability = 0.5 (5/100)
Probability-continuous events
Probability-continuous events are unrepeatable and therefore cannot define probability
This contrasts probability-discrete events where probability can determined between 0-1
What is face validity?
Face validity = does the study look like it’s measuring what the researcher is intending to find?
What is construct validity?
Construct validity = construct validity is important in establishing the overall validity of a method - this type of validity is especially important when what’s being researched can’t be measured/observed directly
What is concurrent validity?
Concurrent validity = how well the result of a test correlates with the results of a different established test
What is predictive validity?
Predictive validity is the measure of how well a test can accurately predict future outcomes based on the psychological construct it measures
What is test-retest reliability?
Test-retest reliability is when the same test is given to the researchers on two seperate occasions, thus allowing correlation between results
What is inter-rater reliability?
Inter-rater reliability = two seperate researchers conduct the same research on seperate occasions - these results are then correlated
What is internal consistency (reliability)?
Internal consistency = measures the correlation between two different versions of a test that are designed to measure the same thing
What is parallel forms (reliability)?
Parallel forms measures the correlation between multiple items in a test that are intended to measure the same thing
What is reliability?
Reliability refers to the degree of consistency. It is measured via a correlation coefficient, such as Pearson’s R. Different types include test-retest, inter-rater, parallel forms, internal consistency
Differentiate internal reliability and external reliability.
Internal reliability = how consistently different items within a test measure the same concept
External reliability = how consistently a test produces similar results over repeated administrations or different conditions
What is internal consistency reliability?
Internal consistency = degree to which different test items (that measure the same concept) yield similar results -> measured via Cronbach’s alpha or Split-Half method (quantifies internal consistency reliability between 0-1)
What is data?
Data is;
-Information in numeric form
-Can represent anything
-Allows mathematical and statistical analysis
-Meaningless without labels and context
Continuous vs Categorical data
Numeric scale of which is being measured = continuous (ordinal, interval) (statistical analysis)
Fixed units = categorical (nominal) (histograms, bar charts etc)
Basic method of collecting data
-Procedure eg questionnaire
-Is there a relationship between the variables being measured?
-Differences between groups? (use of a control group)
-Assessment of the participant pre and post the dependant variable (eg mood before dance, mood after dance) -> difference?
What are the sections that make up a research report?
Sections making up a research report;
Title, abstract, intro, method, results, discussion, references
How science is communicated in psychology
Communicating science; peer reviewed scientific journal articles, formulaic, fully explained, replicable, peer review process, student research reports
Describe the title and abstract sections of a research paper;
Describe the title and abstract sections of a research paper;
-Brief summary of report (<150 words)
-Intro, background, and research Q
-Basic method and experimental design
-Brief results and take-home message
Description of the introduction to a research paper
Research intro;
-Gives specific context and justification for study
-Start by situating the research area with citation
-Explain what research tells us and what it doesn’t tell us
What to include in the method section of a research paper?
Method;
-How many participants?
-Who are the participants?
-Where are the participants from - how they were recruited and pay
What is included within the design section of a research paper?
Design;
-Describes structure of the research, including the variables, conditions and counterbalancing
-Meaningful labels
-Indicate what you’re looking for - correlation, difference, etc?
Stimuli / Measure / Apparatus in research paper
-What did you use?
-Stimuli - recorded sounds, images, spoken or written words
-Relevant characteristics?
-Measures - questionnaires, tests etc
-Apparatus
-Inlcude word lists, images, etc in appendices
Procedure section of a research paper
Procedure;
-Chronological order, clear, consice
-Consider participants viewpoint
-Dont repeat other sections; refer to them
-Explain any relevant differences between conditions
-Brief instructions may be quoted
Results section of a research paper
Results;
-Focus on data, not previous studies
-Give facts -> save interpretation for later in the report
-Appropiate analysis
-Descriptive and inferential statistics
Discussion section of a research paper
Discussion;
-Mirrors intro
-Start with summary of key effects
-Answer to the question intended to adress
-How findings relate to previous work?
-Potential next questions?
-Be critical, analytical, original and thoughtful
References section of a research paper
References;
-APA 7 format
-Every article, book or chapter discussed in the report should appear in references section
-Only sources cites
-Citations should be in correct format
-References listed in alphabetical order by first author
What is descriptive statistics?
Descriptive statistics = techniques for summarising, classifying and describing various aspects of numerical data
What are frequency distributions and some examples?
Frequency distribution represent the number of occurences of each value in a data set - this includes bar charts and histograms
What are the three ways of describing distribution?
-Skewness (deviation from symmetry - same shape on both sides of centre)
-Modality (number of major peaks in a distribution - unimodal or bimodal)
-Kurtosis (measured tailedness - light of heavy)
What are the various measures of central tendency?
Measures of central tendency = numerical values referring to the centre of a distribution
-Mode, median, mean
-Provide information about the location of a distribution
-Affected by outliers/extreme values
Mode value
-Value that occurs most often
-Applicable to categorical and continuous data
-If 2 adjacent scores have same highest frequency: report the average of these values as mode
-Bimodial = dont report mode as a measure of central tendency
-May not represent central tendency at all
Median Value
-Point at which 50% values fall when arranged in numerical order
-If sequence is odd; median is score of individual number (N+1)2
Mean value
-Average (sum of scores/number of scores)
-Basis for inferential statistics
What are the measures of variability?
-How scores are different fro one another
-Range (distance between lowest and highest score in a sample - sensitive to outliers)
-Interquartile range (discards lowest 25% and highest 25%)
Standard Deviation
Standard deviation = how spread out the data is around the mean
-Average distance or deviation
-Easier to interpret than variance
Large SD = data more spread out, high variability
Small SD = data less spread out, low variability