Lecture 10 Flashcards
What is causality/causal relation?
A relationship between cause and effect
e.g. ‘a’ causes ‘b’
In general, a process can have many causes. Those causes are called the ____ for it in its past.
causal factors
An effect can be a cause of many others effects. This is called a ____.
causal chain
Fill in the blanks and name the following condition (cause):
If ‘x’ is a ____ condition (cause) of ‘y’, then the presence of ‘y’ ____ indicates the presence of ‘x’. The presence of ‘x’, however, does not imply that ‘y’ will occur.
Necessary conditions: necessary, necessarily
Fill in the blanks and name the following condition:
If ‘x’ is a ___ condition (cause) of ‘y’, the presence of ‘x’ indicates the presence of ‘y’.
Sufficient: sufficient
What are the specific criteria for inferring causality? (3)
- The cause precedes the effect in time (‘y’ changes after ‘x’: time sequence)
- The two variables are empirically correlated with one another (if ‘x’ changes, ‘y’ changes statistically: correlation)
- The logical relation (is ‘x’ is removed, then ‘y’ disappears: “the observed correlation between the two variables cannot be explained away as the result of the influence of some third variable that causes the two under consideration”)
What is correlation?
The simultaneous change in values of two variables statistically.
e.g. the positive correlation between cigarette smoking and the incidence of lung cancer; the negative correlation between age and normal vision
Many people are confused with the differences between ___ ___ and ___ and make incorrect conclusions.
causal relations; correlations
True or false: a causality must be a correlation.
True
True or false: A correlation must be a causality.
False
True or false: A correlation is a necessary condition, but not a sufficient condition, for causality.
True
True or false: A causality is a sufficient condition for a correlation.
True
Which one is true?
A) a researcher is more likely to find evidence that an effect exists
B) a researcher is more likely to find evidence that an effect does not exist
A
“In many cases, a researcher is more likely to falsely find evidence that an effect exists that to correctly find evidence that it does not”
What term do most studies use to claim correlations in order to proceed cautiously?
Statistically
- ‘a’ is STATISTICALLY associated with ‘b’, rather than ‘a’ CAUSES ‘b’
Why study correlation?
Since correlation is a necessary condition of causality, understanding correlation helps us understand causality in the end.
True or false: To study causal relations, you need to find a lagged process (e.g. ‘y’ lags ‘n’ intervals for time series analysis) and you have to find a strong logical relation between ‘x’ and ‘y’
True
(lag = fixed amount of passing time; one set of observations in a time series is plotted (lagged) against a second, later set of data)
e.g. Change of industrial structure, family size and divorce rate
https://www.khanacademy.org/math/probability/scatterplots-a1/creating-interpreting-scatterplots/v/correlation-and-causality
True or false: It is easy to confirm causality in the real world.
False
How can you control extraneous variables?
Study design
What is internal validity?
The extent to which we can make clear-cut inferences about cause-consequence relations (in the sample being studied)
Name and describe threats to internal validity. (7)
- History; during the course of research, extraneous events may occur that will confound the results
- Maturation or the passage of time; people continuously grow and change
- Testing; the process of testing itself affects the results
- Instrumentation changes; if we use different measures of the dependant variable at posttest than pretest, how can we be sure they are comparable to each other?
- Statistical regression; because we are most likely to begin interventions for human problems that are inherently variable when those problems are are their most severe, we can expect some amelioration of the problem to occur solely because of the natural peaks and valleys in the problem and not necessarily because of the interventions
- Sampling and selection bias; comparisons don’t have any real meaning unless the groups being compared are really comparable
- Ambiguity about the direction of causal influence; there is a possibility of ambiguity concerning the time order of the independent and dependent variables
Name this type of test, and describe its limitations:
X O
One-shot case study; it doesn’t even establish correlation, and it fails to control for any threats to internal validity
Name this type of test and its limitations:
O1 X O2
One-group pretest-posttest; it does not account for factors other than the independent variable that might have caused the change between pretest and posttest results–factors usually associated with the following threats to internal validity: history, maturation, testing, and statistical regression.
Name this type of test and its limitations:
X O
O
Posttest only with non-equivalent groups; fails to control for the threat of selection bias–without pretests, we have no way of knowing whether the score of the two groups would have differed as much to begin with.
Some studies using pre-experimental designs can be valuable despite their extremely limited degree of internal validity. What tends to make them valuable as pilot studies? (4)
- To generate tentative exploratory or descriptive information regarding a new intervention about which little is known
- To learn whether it is feasible to provide the new intervention as intended
- To identify obstacles in carrying out methodological aspects of a more internally valid design that is planned for the future
- To see if the hypothesis for a more rigorous study remains plausible based on the pilot study results
