Final Flashcards
The entire group of people about which we
wish to generalize
Population
A portion of a population
Sample
Sampling only those who are easy to contact
Convenience sampling
Sampling only those who volunteer
Self-selection
What are the 4 non-probability sampling techniques?
Convenience sampling
Quota sampling
Purposive sampling
Snowball sampling
What are the 4 probability sampling techniques?
Simple random sampling
Systematic sampling
Stratified random sampling
Cluster sampling
Every individual in the population
has an equal chance of being selected
Simple random sample
Sample is selected according to a random starting point and a fixed periodic interval
Systematic sampling
Strata are formed based on members’
shared attributes or characteristics
Stratified random sampling
Uses “natural” but relatively heterogeneous
groupings in a population.
Cluster sampling
An extension of convenience sampling, based on the characteristics of the sample and the purpose of the research
Purposive sampling
Participants are chosen out of specific subgroups that are identified, with convenience sampling used to select the required number of participants from each subgroup
Quota sampling
Participants recruit other participants, used to collect data when the desired sample characteristic is rare, or it is difficult to locate respondents
Snowball sampling
Describes the data (variables) quantitatively
Descriptive statistics
What are the differences between parameters and statistics?
Statistics describe samples, parameters describe populations
A spreadsheet of our variables and their values
Data matrix
A table which provides the number of or
frequency of each possible value
Frequency distribution
A way of providing a graphical representation of the frequency of one variable of interest
Histogram or dot plot
Measure of central tendency that can tell us where most of our scores in our dataset center around
Mean
The middle score that splits the dataset in half
Median
The most common number in a dataset
Mode
What are the two ways to measure spread/variability in data?
Variance and standard deviation
The average spread that each number in our dataset has around the mean
Variance
The square root of the variance which provides a benchmark or indicator of spread for our dataset
Standard deviation
Specify how far away (in standard deviation units) one score is from the mean
Z-score
How does one calculate a z-score?
Difference between the individual score and the mean divided by the standard deviation
A degree of how one variable changes in
relation to the other variable
Covariance
A standardized covariance ranging from -1.0 to +1.0
Pearson’s r
Tell us the strength of a relationship between two variables
Effect sizes
Tells us the distance between the means of two groups in standard deviation units
Cohen’s d
A set of procedures that use the rules of probability to make inferences or generalizations about a population using sample data
Inferential statistics
What are three types of point estimates?
Percentage
Effect size
Strength of relationship
Range around the point estimate that often contains the true value (population value)
Confidence interval
What does the CI for a percentage estimate include?
Percent estimate +/- margin of error
On which three factors does the margin of error vary?
SD
Sample size
Level of confidence
What is the difference between CI and NHST?
CI represents a “new statistics” based on estimation
Null hypothesis significance testing is binary (yes/no)
What do H0 and H1 imply?
H0: Null hypothesis, no effect
H1: Alternate hypothesis, significant treatment effect
Describe the difference between a type l and type ll error
Type l: False positive, reject null hypothesis when it is true
Type ll: False negative, fail to reject null hypothesis when it is false
What hypothesis tests are appropriate for determining the mean difference for two groups?
Independent samples t-test, dependent samples t-test
What characteristics define independent samples t-tests?
Independent groups, between-subjects design
What characteristics define dependent-sample t-tests?
Dependent groups, within-subjects design
What hypothesis tests are appropriate for determining mean differences for two or more groups?
One-way ANOVA, repeated-measures ANOVA, two-way ANOVA
What is the f-ratio?
Between groups variability over within groups variability
What hypothesis test is appropriate for assessing the relationship between two numerical variables?
Pearson correlation: Correlation coefficient r and measure of effect r squared
Printed instruments that the respondents complete
Paper and pencil questionnaires
Online surveys created as Web forms with a database to store the answers
Electronic (e-surveys)
Respondents complete questionnaires that are completed on paper and returned via mail
Mail surveys
Collecting data using a telephone to contact respondents
Telephone surveys
Collecting data using a variation
of the different survey methods
Mixed-mode surveys
Collecting data face to face
Interviews
Questions that allow people to provide detailed answers.
Open-ended questions
Questions provide a limited selection of available responses.
Forced-choice questions
Questions use a rating scale to indicate level of agreement
Likert scale
Questions provide a numeric scale anchored by adjectives
Semantic differential
Which types of questions are appropriate for nominal measurements?
Dichotomous, demographic, forced choice
What types of questions are appropriate for ordinal measurements?
Rank order, scales, and likert-type
What types of questions are appropriate for interval measurements?
Semantic differential
Questions that require answers for two different options, leading to confusion
Double-barreled
Questions worded in such way that they imply a derogatory association
Negatively-worded questions
Responding in one set type of pattern
Response sets
Responding “yes” to all questions
Acquiescence
Choosing the moderate option
Fence sitting
Responding in a way that makes you look good
Socially desirable responding/faking good
The extent to which a measure is repeatable or stable; it refers to the consistency of a measure
Reliability
The extent to which the survey measures the construct we want to measure and no other related constructs
Validity
Refers to the degree to which
test results are consistent over time
Test-retest relibility
Respondents’ answers are compared on slightly different versions of a survey designed to measure the same construct
Parallel forms
Refers to the degree that the survey items are measuring the same construct
Internal consistency
What is the most common measure of internal consistency?
Cronbach’s alpha
A measure of consistency wherein a survey is split into two equal parts and the results for each half are compared with one another
Split-half reliability
The extent to which a survey is subjectively viewed as measuring the concept
Face validity
Refers to the extent to which a measure represents all features of a given construct and focuses on the content of the survey items
Content validity
What are the two types of content validity?
Convergent and discriminant
The extent to which a measure is related to an outcome. It involves comparing our survey results with other measures or outcomes already considered valid.
Criterion-related validity
The extent to which a survey actually measures the construct it intended to measure and focuses on whether our survey measures a construct that cannot be directly observed.
Construct validity
What are the four types of observational research?
Naturalistic observation
Participant observation
Structured observation
Field experiment
Form of observation in which the observer is passive and does not intervene.
Naturalistic observation
Form of observation through which the observer can access normally “closed” situations (disguised on undisguised)
Participant observation
Form of observation used to observe behaviours that are difficult to see naturally
Structured observation
Manipulation of variables as in a true experiment that occurs in natural settings
Field experiment
When observers see what they expect to see
Observer bias
When participants confirm observer
expectations
Observer effects
What is one key way to minimize observer bias and effects?
Using a masked design
What are the three solutions to reactivity during observations?
Blend in, wait it out, or measure the behaviour’s results
When participants react to being watched
Reactivity
When is it ethical to observe the behaviours of others without their consent?
When the behaviours take place in an environment that does not have an expectation of privacy
Associations that involve exactly two variables
Bivariate correlations
What makes a study correlational?
Having two measured variables and no manipulated variables
When interrogating association claims, what does construct validity refer to?
How well each variable was measured
When interrogating association claims, what does statistical validity refer to?
How well the data supports the conclusion
When interrogating association claims, what does external validity refer to?
Who the association can be generalized to
When interrogating association claims, what does internal validity refer to?
Whether or not a causal inference can be made from the association
How do outliers affect associations?
Skew the data to appear less associated than is likely true
How does restricting the range affect associations?
The degree of the association might appear minimized
What are the three causal criteria?
Covariance, temporal precedence, internal validity
Designs that involve more than two measured variables
Multivariate designs
Which causal criteria do longitudinal designs help address?
Temporal precedence
Which causal criteria do multiple regression analyses help address?
Internal validity
Another variable is generating the association between two other variables
Mediator
Another variable that controls the degree of association between two variables
Moderator
Influences that interfere with an accurate measurement between the independent and dependent variable
Confounds
What are two advantages of within-groups designs?
Participants are equivalent, require fewer participants
What is one strategy to avoid order effects?
Counterbalancing
When being exposed to one condition affects how participants respond to other conditions
Order effects
What are three disadvantages of within-groups designs?
Potential for order effects
Might not be practical or possible
Demand characteristics
Experiencing all levels of the independent variable (IV) changes the way participants act
Demand characteristics
The effect of one independent variable depends on the level of the other independent variable
Interaction
Design with two or more independent variables
Factorial
Differences between the levels of one independent variable across levels of the other independent variable
Main effect
What is more important: Interactions or main effects?
Interactions
When interrogating causal claims, what does construct validity refer to?
How well the variables were measured and manipulated
When interrogating causal claims, what does external validity refer to?
Who or what the causal claim can generalize to
When interrogating causal claims, what does statistical validity refer to?
How much, how precise, what else is known
When interrogating causal claims, what does internal validity refer to?
If there are any alternative explanations for the results
What are the six potential threats to internal validity in one-group or pretest-posttest designs?
Maturation threats
History threats
Regression threats
Attrition threats
Testing threats
Instrumentation threats
What are the three potential threats to internal validity in any study?
Observer bias
Demand characteristics
Placebo effects
Which three things could be responsible for a null effect?
Really no difference
Not enough between-group difference
Within-group variability obscured the group differences
When are quasi-experiments appropriate?
When researchers do not have full experimental control
What are the four types of quasi-experiments?
Nonequivalent control group posttest only
Nonequivalent control group pretest/posttest
Interrupted time-series
Nonequivalent control group interrupted time-series
Quasi-independent variable with dependent variable measured only once after exposure to the IV
Nonequivalent control group posttest-only
Quasi-independent variable with dependent variable measured once before and once after exposure to the IV
Nonequivalent control group pretest/posttest
A variable is measured before and after an “interruption”
Interrupted time-series
Two quasi-independent variables (group/condition, time) and a dependent variable
Nonequivalent control group interrupted time-series
What are three similarities between correlational studies and quasi-experiments?
Both may use independent-groups designs.
Neither use random assignment.
Neither use manipulated variables.
Designs in which only a few individuals are studied
Small-N designs
Small-N design in which baseline is assessed followed by introducing the intervention
Stable-baseline designs
Small-N design in which different baselines are assessed in relation to the effect of the intervention
Multiple-baseline designs
Small-N design in which the treatment is introduced and then taken away
Reversal designs
What are the major themes of qualitative research?
Answers in-depth social questions about “how” and “why”
Holistic, formative, thematic
Emphasis on studying things in their natural environment
Uses smaller sample sizes and more flexibility in sampling
What are the three major qualitative theories?
Grounded theory
Ethnography
Phenomenology
Qualitative comparative method that constructs theory from the process itself (inductive)
Grounded theory
Qualitative method that systematically studies patterns between people and cultures
Ethnography
Qualitative method that rejects data and themes altogether and collects thoughts and objects which influence each other
Phenomenology
What are three common methods of data collection for qualitative research?
Interviews
Focus groups
Participant observation
List of thematic codes with their definitions and several examples of what could be included and not included under this heading
Codebook
What are the three types of replication?
Direct replication
Conceptual replication
Replication-plus-extension
Exact replication of a study
Direct replication
Replication that maintains the same concept but changes the way it’s operationalized/measured
Conceptual replication
Replication in which the original experimental design is maintained with an added component
Replication-plus-extension
Quantitative technique that calculates effect size across studies
Meta-analysis
What are four questionable research practices?
Underreporting null findings
HARKing
p-hacking
Using small samples
What does HARKing refer to?
Hypothesizing after the results are known
What are three examples of transparent research practices?
Open science
Preregistration
Encouraging large samples
An experimental group improves over time only because of natural development or spontaneous improvement
Maturation threat
An experimental group changes over time because of an external factor that affects all or most members of the group.
History threat
An experimental group whose average is extremely low (or high) at pretest will get better (or worse) over time because the random events that caused the extreme pretest scores do not recur the same way at posttest
Regression to the mean
An experimental group changes over time, but only because the most extreme cases have systematically dropped out and their scores are not included in the posttest
Attrition threat
A type of order effect: An experimental group changes over time because repeated testing has affected the participants. Practice effects (fatigue effects) are one subtype.
Testing threat