Lecture 3 - Understanding Variables and Variation Flashcards
what is the main problem with assigning participants to groups in between-groups experiments?
the main problem is that groups will still differ by chance, usually just a bit, but sometimes a lot. This can be addressed by random assignment and using statistical analysis to assess how often chance alone would produce a result as strong as the one observed
what are order effects in within-groups experiments, and how can they be controlled?
order effects occur when the order in which treatments are given affects the results. This can be controlled by order counterbalancing, where equal numbers of participants experience each possible order of treatments
describe the basic logic of a simple idealized experiment
the basic logic involves setting up two situations that are identical, changing the state of the supposed causal variable in one of the situations, and measuring the outcome in both situations. If the outcome differs between situations, the supposed causal variable must be a cause; if the outcome doesn’t differ, it must not be
what is a confound in the context of experimental research?
a confound is a specific variable, other than the supposed causal variable of interest, the presents an alternative explanation for important results
what are some practical problems with experiments?
practical problems with experiments include reduced realism, high costs and time consumption, and ethical issues such as violating consent or causing suffering
how can confounds be addressed in between-groups and within-groups experiments?
In between-groups experiments, confounds can be addressed by randomly assigning participants to groups. In within-groups experiments, confounds such as order effects can be addressed by order counterbalancing
define the term “construct” in psychological research
a construct is a concept of something of interest in a study that may or may not actually exist. It is not directly observable, and its state is uncertain and must be judged from variables
what are operations in the context of research methodology?
Operations are methods of changing or assessing the state of a construct. There are two kinds: manipulations (changing the state of a construct) and measures (assessing the state of a construct)
what is the difference between independent and dependent variables?
independent variables are considered causes of other variables, while dependent variables are considered to be affected by other variables
how can variation in a single variable be described using histograms?
Variation in a single variable can be described by counting how many times each of its possible states occurred in the data. This is represented visually in a histogram.
explain the information given by the four scales of measurement with examples
the four scales of measurement are:
- nominal: Names of variable’s states (e.g. political party affiliation)
- ordinal: ranking (order) (e.g. place in a race)
- interval: Distance between states (e.g. temperature in degrees)
- ratio: proportion distance from a true zero (e.g. net worth in dollars)
explain the use of histograms, scatterplots, and box and whisker plots in describing variation
histograms are used to show the frequency of different states of a variable
scatterplots are used to show relationships between two variables, and box-and-whisker plots summarize data distribution, showing the median, quartiles and possible outliers
Is measuring driving speeds on several roads before and after the random deployment of speed cameras an experimental or observational study?
It is an experimental study because it involves the random deployment of speed cameras, which is an intervention
is recording driving speeds on stretches of road near and far from already-deployed speed cameras an experimental or observational study?
It is an observational study because it measures driving speeds without intervening to change the state of any variables
what are some qualities of good theories in research?
Qualities of good theories include falsifiability, predictive power, scope, simplicity and fruitfulness
