Midterm #2 Flashcards
Target population
defined by researcher’s specific interests
Accessible population
Portion of population who are accessible to be recruited for the study
Sample population
relatively smaller group of individuals who participate in the study
Simple random sampling (random)
- Randomly select participants from list containing total population
- Each individual has equal and independent chance of selection
Systematic sampling (random)
- Select every nth participant from list containing total population after random start
Stratified random sampling (random)
- Divide population into subgroups and randomly select equal numbers from each subgroup
Proportionate stratified random sampling (random)
- Divide population into subgroups and randomly select from each subgroup so proportions in sample correspond to proportions in population
Cluster sampling (random)
- Randomly select clusters from a list of all the clusters in the population
Convenience sampling (nonrandom)
- Select individual participants who are easy to get
Quota sampling (nonrandom)
- Identify subgroups to be included, then establish quotas for individuals to be selected through covenience
Descriptive
Intended to answer questions about the current state of individual variables for a specific group of individuals. (Describe specific characteristics of a specific group of individuals)
Correlational
- Measure two variables of interest for each individual
- Look at data graphically - scatterplot
Experimental
- Intended to answer cause-effect questions about the relationship between variables
- rigorous control to help ensure an unambiguous demonstration of the cause-effect relationship
Quasi-experimental
- Attempts to answer cause-effect questions about relationship between two variables, but answers tend to be ambiguous.
Nonexperimental
- Demonstrates relationship between variables without explaining relationship.
- Does not use rigour and control or produce cause-effect explanations.
External validity
- The extent to which the results obtained in a research study can hold true outside that specific study.
Internal validity
- Changes in one variable are followed by changes in another variable and no other variable provides an alternative explanation for the results.
Experimenter bias (threat to validity)
- the findings of the of a study are influenced by the experimenter’s expectations or personal beliefs about the study’s outcome.
Reactivity (threat to validity)
- participants modify their natural behaviour in response to the fact that they are aware they are being studied
- behaviour can change by being overly cooperative or defensive/uncooperative.
Confounding variable (threats to validity)
- extraneous variable (usually unmonitored)
- changes systematically along with the two variables being studied.
- alternative explanation for observed relationships between the two variables.
Assignment bias (threats to validity)
Occurs when the process used to assign different participants to different treatments produces groups of individuals with noticeably different characteristics
Selection bias (threats to validity)
- Sampling procedure favours the selection of some individuals over others.
- If the sample doesn’t accurately represent the population, the results will not generalize to the population
Manipulation (experimental)
Manipulate one variable to create two different treatment conditions.
- Third variable
- Directionality
Measure (experimental)
Measure a second variable to obtain a set of scores in each treatment condition
Comparison (experimental)
Compare the scores in treatment A with the scores in treatment B
Control (experimental)
- The second distinguishing characteristic of an experiment is control of extraneous variables
- Ensure observed relationship is not contaminated by the influence of other variables.
Threats to validity (experimental)
- Assignment bias (different participants to different treatment groups with noticeably different characteristics)
Mitigating threats to validity
Holding variables constant
- making variables the same for every observation
- environmental variables: standardize the environment and procedures
- individual differences variables: hold demographic variables constant
Matching
- balance variables in each treatment condition
Random assignment
- passive control technique
- disrupting systematic relation
- unpredictable and unbiased procedure to distribute different values of each extraneous variable across the treatment conditions
Manipulation check
- included in the study to measure whether the independent variable had the intended effect on the participant
Placebo effect
Believed to be psychosomatic: the mind (psyche) has an effect on the body (somatic)
- the individual thinking/believing it’s effective, causes a response to the medication
Placebo effect
Inactive drugs, Nonspecific therapy, Non alcoholic beverages
Believed to be psychosomatic: the mind (psyche) has an effect on the body (somatic)
- the individual thinking/believing it’s effective, causes a response to the medication
What is a between subjects design?
- obtain each of the different groups of scores from a separate group of participants
- comparing different groups of individuals
- individual scores
Advantages of subjects design?
- Each individual score is independent from the other scores
- Can be used for a wide variety of research questions
Disadvantages of subjects design?
- Require a relatively large number of participants
- Individual differences
Specific threats to validity (between subjects design)
- Differential attrition: when participants withdrawal from a study before it’s completed
- Communication: when participants from different conditions talk to each other…
(Diffusion) - treatment effect spread between groups
(Compensatory equalization) - untreated groups demand equal treatment
(Compensatory rivalry) - untreated group works hard to show they can perform just as good
(Resentful demoralization) - untreated group becomes less productive and motivated
What does the t-statistic represent?
measures the size of the difference relative to the variation in your sample data
