Kin209 Midterm 2 Flashcards
Causal claims
Post-positivism
Based on the notion that there is a single reality, or a single objective truth that is waiting to be discovered through research.
Key defining features:
Determinism: causes determine effects (use of experimentation)
Reductionism: ideas can be reduced to small, discrete, and testable set (ie variables that comprise hypotheses and research questions)
Empirical observation and measurement: knowledge is based on careful observation and measurement
Theory testing: theories are tested corroborated, or refined
Determinism and causality defined
Determinism: the philosophical doctrine that assumes that every event has causes
-events are determined, or caused to happen; they are not ‘free’ in any other way
-injury prevention vs “accidents”
Causality: the principle that everything has a cause
-the relation of cause and effect
Causation
Effects can be the result of multiple causes
To claim causation:
1) covariation (correlation) between variables;
2) the relationship is not found to be the result of some third variable (isolation)
3) effects follow causes (temporal precedence)
4) the manipulation of a cause will result in the manipulation of an effect
Three criteria for establishing causation between Variable A and B
covariance (correlation): the study's show that as A changes, B changes (high levels correlate with high, vice versa) temporal precedence (cause & effect): the study's method ensures that A comes first in time, before B internal validity: the study's method ensures that there are no plausible alternative explanations for the change in B; A is the only thing that changed. (accounts for a third variable (C) or confounder that implies that manipulation of A will affect B)
Goals of science revisited
1) describe: using language to describe results that are scientifically a reflection of relationships should be avoided
- this is the way things are, cannot claim causation
2) explain-causation, how
3) predict-causation, why
Describing the type of claims
causal: -affects -leads to -reduces -has an effect on -experimental study Association claim: -associated -related to -linked to -correlated frequency claim -counts: x number of, x percent of, x-y per day
Frequency claims
Describe a particular rate or degree of a single variable
-how many, how often? (40% of Canadian adults are physically active)
Involve only ONE measured variable
Association Claims (Morling)
Argue that one level of a variable is likely associated with a particular level of another variable
-involve at least two measured variables (unlike frequency claims that focus on one)
Variables that are associated are CORRELATED
-positive
-negative
-zero
No attempt to manipulate, control, or interfere with variables
Correlation does NOT equal causation
Ice cream and shark attacks
Making predictions based on associations (Morling)
- some association claims are useful because they help us make predictions
- the stronger the association between the two variables, the more accurate the prediction
- both positive and negative associations can help us make predictions, but zero association cannot
Causal claims (Morling)
One of the variables is responsible for changing the other (cause and effect)
-one measured variable (DV)
-one manipulated variable (IV)
Typically takes the form of an experimental design
-experimental group, control group
Mediator (MacKinnon)
-A third variable that is intermediate in the causal chain relating X and Y
In a mediation model, the IV causes the mediator which then causes the DV
-X,Y,Z
-no relationship between x and y without
Moderator (MacKinnon)
A variable that changes the sign or strength of the effect of an IV on a DV
- typically an interaction such that the effect of the IV on the DV depends on the level of the moderator variable
- effect modifiers (gender)
Mediators and moderators
mediating variable explains the relationship between two or more other variables
environment->behaviour->health
moderator enhances or modifies the relationship between two or more other variables
behaviour->Health
SES
-with whom will you conduct your intention
Confounder
A variable that changes the relationship between an independent variable (IV) and dependant variable (DV) because it is related to both the IV and the DV
Though it explains/helps understand the relationship between X and Y because it is related to both, a confounder is not part of the causal mediation process
(smoking is related to cancer and alcoholism is related to both, but alcohol is not part of the causal chain between smoking and cancer)
a couple more confounder
germs, bad smells, disease
Internal and external validity
validity buffet
Validity of quantitative research designs -internal validity -external validity Validity of measures -content validity -statistical validity
Internal validity
- The ability to claim that any changes in an outcome (DV) is the result of a treatment or intervention (manipulation of IV), and not a result of other factors related to the sample, the measures, research techniques, or other potential threats
- Can the outcome be attributed to only the intervention or could something else have caused it? (consider: confounding variables)
Threats to internal validity
Threats to the experimental procedure: -testing -instrument accuracy Threats to treatment or manipulation -diffusion of treatment -halo effect (researcher bias) Threats to the participants: -maturation -history -regression (to the mean) -selection bias (volunteers) -experimental drop-out -placebo effect -Hawthorne effect (act in favourable ways because they are being observed)
Threats to experimental procedure
Testing: participants may become familiar with the test, which could influence future performance (VO2, pedometer)
Instrument accuracy: Measures must be valid and reliable, in good working order, and strive to be free of calibration error, inappropriate use, difference in collection technique, etc.
Threats to treatment or manipulation
Diffusion of treatment: participants in one group talk about their experiences with participants in another group
Halo effect: when researchers have some expectation about the performance of a participant and are in a position to assess performance, their knowledge of experimental conditions could influence their judgement on outcomes or could influence the way they respond to participants (blinded vs double blinded)
Threats related to the participants
Maturation: change may naturally happen to participants over time via growth, learning, or maturation
History: events other than the experimental treatment impact of the study (covid with school research)
Regression: extreme scores have a tendency to regress back toward the mean. Scores that naturally change in the direction of the mean (extreme highs and lows)
