final exam Flashcards
define: non-experimental research
a research strategy that involves determining whether an association exists between variables by comparing two or more groups of participants that represent different values of one variable to see if the groups also differ on another variable
define: quasi-independent variable
a predictor variable in non-experimental or quasi-experimental research
define: quasi-experimental research
A research strategy that involves varying the levels of one variable and measuring another variable, but that lacks the control needed to establish a cause and effect association between variables
true or false: quasi-experimental research involves more control/manipulation than non-experimental
true
the term non-experimental research specifically refers to research looking at the difference between ____ ____ of two or more groups of participants
mean scores
why is the association between variables that has been established through non-experimental group comparisons sometimes described as correlational research?
because of lack of control
true or false: some sources categorize quasi-experimental, correlational, and descriptive research all as types of non-experimental research
true
what types of designs can refer to both non-experimental and experimental group designs?
- between subjects designs
- within subjects designs
what types of statistical analysis can be used for between subjects and within subjects designs?
- t-test
- ANOVA (analysis of variance)
define: non-equivalent group designs
a study in which the different groups of participants are not randomly assigned to groups and are therefore considered not to be equivalent
what is a threat to internal validity in non-equivalent group designs?
individual differences
what are the 3 types of non-equivalent group designs and define them
- differential research designs = a study that compares a dependent variable between two or more pre-existing groups of participants (ex: self-esteem and academic performance - cannot randomly assign people to have different levels of self-esteem)
- Post-test only non-equivalent control group design = a study that compares a dependent variable (outcome) between a group of participants who received a treatment and a group of participants who did not receive treatment
- Pretest-posttest non-equivalent control group design = a study that compares a dependent variable (outcome) between a group of participants who received a treatment and a group of participants who did not receive a treatment
what does a pre-test control?
time related variables
what type of design is a non-equivalent group design?
a between subjects non-experimental group design
what is a within-subjects non-experimental group design?
pre-post design = a study that compares a dependent variable (outcome) measured in one group of participants before and after they receive an intervention (no control group is used)
why is a pre-post design considered a within-subjects experimental group design?
because it’s the same group of participants who is measured on the dependent variable two (or more) separate times
what are the three types of pre-post designs and define them
- Pretest-posttest design = each individual is measured only once before and once after treatment
- Time-series design = each participant is measured multiple times before and after treatment
- Interrupted time-series design = the “intervention” is not manipulated by the researcher
define: developmental research designs
Research designs used to evaluate changes that occur as people get older
what are the 2 types of developmental research designs and define them
- Cross-sectional design = a research design in which people of different ages are compared at the same point in time
- Longitudinal study design = a research design in which the same participants are studied over time using multiple measurements - individual differences do not provide an alternative explanation for results
another term for an independent variable is a ___
factor
define: factorial design
a research design that includes two or more independent variables (factors)
define: two-factor design
a study with two independent variables
define: higher-order factorial design
a study with three or more independent variables
why do we use factorial designs?
- Humans are complicated and influenced by a variety of factors
- Allows us to look at factors acting together that uniquely affect behaviour in ways that are different from either factor acting alone
what is the simplest form of a factorial design?
2 independent variables with 2 levels = 4 different study conditions with their own means
what is a 2 factor design?
2 levels of the first independent variable times 3 levels of the second independent variable
what is a higher-order factorial design?
2 levels of the first independent variable times 3 levels of the second independent variable times 2 levels of the third variable
what does the statistical analysis involve for factorial designs?
looking at the mean DV scores when considering each IV alone (main effects) and also mean DV scores for each level of each IV combination with each level of the other IVs (interactions) using an ANOVA test
what type of p value are we looking for in factorial designs?
less than 0.5
define: collapsed across
just comparing one independent variable at a time and pretending the other does not exist
true or false: if a graph has parallel lines, there is an interaction
false
true or false: if a graph does not have parallel lines, there is no interaction
false
what is a mixed design?
Mixed design = a factorial study that includes at least one between-subjects factor (exposed to only one level of the IV) and at least one within subjects factor (exposed to more than one level of the IV)
define: combined strategy
a factorial study that includes at least one true experimental independent variable (random assignment to IV level) and one quasi-independent variable (assignment to IV level based on a pre-existing characteristic)
define: correlational designs
Looking at the association between 2 variables at a time
true or false: when the correlation is 0, there is no association between the variables
true
true or false: as the correlation gets closer to 1, the correlation gets stronger
true
what can correlations demonstrate?
- Alternate-forms reliability
- Internal consistency
- Concurrent validity
- Convergent & divergent validity
- predictions
- predictive validity
- academic success
define: regression
A statistical process of finding a linear equation that produces the most accurate predicted values of a dependent variable (Y) using a predictor variable (X)
define: multiple regression
involves two or more predictor variables
what is the regression equation and what do the symbols mean?
Y = bX + a
- Y = a specific value of the dependent variable
- X = a specific value of the independent (predictor) variable
- Slope (b) = how much Y changes when X increases by 1
- Y-intercept (a) = the value at which the regression line intersects the Y axis
what are the limitations of correlational designs?
- Can’t determine causal associations
- Directionality problem
- Third variable problem
define: descriptive research
a research strategy that involves measuring variables as they currently exist, without manipulations, comparisons or evaluations or their associations with other variables
how is descriptive research usually presented?
through numerical data like averages, means and percentages
what are the different types of descriptive research?
- survey research
- observational research
- case studies
what are 2 issues with behavioural observations as measurements of variables?
- The natural behaviour of participants can be disrupted by the presence of the researcher (or recording equipment)
- Observational measurements may be more subjective because they rely on the interpretations of the observer
what are three recommendations for ensuring objectivity in behavioural observations?
- Develop a list of well-defined behaviour categories
- Use well-trained observers
- Use multiple observers to establish inter-rater reliability (at least with a sample of observations)
what are the three types of behavioural observation?
- Naturalistic observation = as unobtrusively as possible (without being noticed), researchers directly observe and record the naturally-occurring behaviour of participants as they engage in their real-life activities
- Participant observation = researchers directly observe and record behaviour of participants in their real-life settings while engaging in the same activities as the participants
- Contrived observation = researchers directly observe and record the behaviour of participants in environments or situations that are specifically arranged to facilitate the occurrence of the behaviour(s) of interest
what are the three ways to quantify behavioural observations?
- Frequency method = the observer counts the number of occurrences that each behaviour occurs within a specific time period
- Duration method = the observer records how much time a participant spend engaging in a particular behaviour
- Interval method = the observation period (e.g. 30 minutes) is divided into shorter intervals (e.g. 10 seconds) and the observer records whether or not the behaviour occurred during each interval
define: single subject experimental designs
Involve the manipulation of an independent variable and control of extraneous variables to infer the existence of a cause-and-effect association using measurements for a single participant
how are single subject designs different from case studies?
case-studies do not necessarily involve the manipulation or control of any variables so they are not experiments
define: phase
a series of consecutive observations of the same individual under the same conditions
define: baseline phase
a series of observations made before any treatment is administered (identified by the letter A)
define: treatment phase
a series of observations made while the treatment is being administered (identified by the letter B)
define: ABAB design
a single-subject experimental design consisting of four phases:
1. Baseline
2. Treatment
3. Return-to-baseline
4. Second treatment phase (aka reversal design)
what are three ways to control extraneous variables in ABAB designs?
- Baseline measurements are taken before the intervention starts
- Repeated observations are obtained during each treatment condition to ensure that factors other than the treatment are ruled out
- Replication - the change between the baseline and treatment conditions must be demonstrated a second time
what are some examples of meaningful changes in ABAB designs?
- Change in average level
- Immediate change in level
- Change in trend (up or down)
- A longer latency of change (if the behaviour eventually changes but does not change right away), it suggests that the change is less likely to be due to the treatment
define: multiple baseline design
a study that begins with two or more baseline phases (observations made before any intervention begins), then a treatment is initiated under one set of conditions while the baseline observations continue for the second condition, then after a short period of time the treatment is initiated in the second condition
define: Multiple-baseline across subjects
each baseline corresponds to the same behaviour for more than one participant
define: Multiple-baseline across behaviours
each baseline corresponds to two distinct behaviours for the same participant
define: Multiple-baseline across environments
each baseline corresponds to the same behaviour but in different environments
define the term significant in regard to nonexperimental and quasi-experimental research designs
means that it is very unlikely that the difference between the groups of scores would occur if there were no corresponding difference in the population
how are non-experimental and quasi-experimental designs set up?
the groups are usually defined in terms of a specific participant variable (e.g., college graduate/no college) or in terms of time (e.g., before and after treatment). These two methods of defining groups produce two general categories of nonexperimental and quasi-experimental designs:
Between-subjects designs, also known as nonequivalent group designs
Within-subjects designs, also known as pre–post designs
what is an example of a non-experimental design study?
A study comparing self-esteem scores for children with a learning disability versus scores for children without a learning disability
what is the distinction between correlational research and differential research?
- In differential research, participant differences in one variable are used to create separate groups, and measurements of the second variable are made within each group. The researcher then compares the measurements for one group with the measurements for another group, typically looking at mean differences between groups
- A correlational study, on the other hand, treats all the participants as a single group and simply measures the two variables for each individual