Psyc 217

Front 1

What are the three core ethical principles for research with human participants as outlined by TCPS?

Back 1

- To show respect for persons (meaning researchers must respect the autonomy of research participants and to protect those who have developing, impaired or diminished autonomy)

- To show concern for welfare (meaning researchers must attempt to minimize risks associated with participating in research, while maximizing the benefits or that research to individual participants and to society)

- To show justice (researchers must treat people fairly and equitably by distributing the benefits and burdens of participating in research)

Front 2

List and describe some of the potential risks and benefits of research.

Back 2

Benefits: education about the scientific process, acquisition of a new skill, or treatment for a psychological or medical problem.

Risks: physical and/or psychological

Front 3

What is informed consent? Why is this used despite the difficulty with obtaining it?

Back 3

Definition: In research ethics, the principle that participants in an experiment be informed in advance of all aspects of the research that might influence their decision to participate.

- This is used to ensure that the participants are aware of all of the risks associated and benefits associated with the research being performed

Front 4

What are the three "R's" to minimize harm to animals in research?

Back 4

1. Replacement: involves finding alternatives to using animals

2. Reduction: process of adopting techniques that require the fewest animals possible

3. Refinement: process of creating procedures that minimize distress

Front 5

What is the difference between deception by omission and deception by commission?

Back 5

Deception by omission: leaving out some of the details

Deception by commission: lying about some of the details

Front 6

Explain additional ethical principles that need to be followed in research.

Back 6

- Don't plagiarize (take someone else's words and/or ideas as your own) however instead paraphrase

- APA/CPA have general ethical principles for conduct in research

Front 7

Describe different ways of constructing survey questions.

Back 7

You can construct surveys based on either open or closed ended questions.

Types of closed-ended questions: Rating scales (Likert or Semantic differential scale)

Front 8

Explain factors to consider when constructing survey questions.

Back 8

- Ensure that question wording is appropriate

- Don't use loaded questions, double barreled questions or negatively worded questions

- Ensure that question order is appropriate

- Both factors influence the construct validity of a survey

Front 9

What does a participant need to do to generate a response?

Back 9

1. Question interpretation

2. Information retrieval

3. Judgement formation

4. Response formatting

5. Response editing

Front 10

What is counter balancing?

Back 10

A method of controlling for order effects in a repeated measures design by either including all orders of treatment presentation or randomly determining the order for each participant.

Front 11

What is a response set and describe issues surrounding analyzing survey data?

Back 11

Definition: when people respond in the same way to all the questions regardless of their content

Problems with response sets:

- Acquiescence: participants end up agreeing with all items no matter what they say

- Fence sitting: participants select an answer in the middle of the scale for questions that are controversial

- Socially desirable responding: participants respond in a way that makes them seem to look better

Front 12

Explain the relationship between sample size and precision.

Back 12

A larger sample size reduces measurement error, and therefore reduce the size of the confidence interval. Although the size of the interval is determined by several factors, sample size is key.

Front 13

Compare and contrast three kinds of probability sampling techniques.

Back 13

1. Simple random sampling: A sampling procedure in which each member of the population has an equal probability of being included in the same

2. Stratified random sampling: A sampling procedure in which the population is divided into strata followed by random sampling from each stratium.

3. Cluster sampling: A sampling procedure in which clusters of individuals are identified. Clusters are sampled, and then all individuals in each cluster are included in the sample

Front 14

Compare and contrast three kinds of non-probability sampling techniques.

Back 14

1. Convenience sampling: selecting participants in a haphazard manner, usually on the basis of availability and not with regard to having a representative sample of the population (aka: haphazard sampling)

2. Purposive sampling: A type of convenience sampling procedure conducted to obtain predetermined types of individuals for the sample

3. Quota sampling: The sample is chosen to reflect the numerical composition of various subgroups in the population. Convenience sampling is used to obtain the sample.

Front 15

Define a factorial design and discuss reasons why a research would use this design.

Back 15

Definition: A design in which all levels of each independent variable are combined with all levels of the other independent variables. A factorial design allows investigation of the separate main effects and interactions of two of more independent variables.

- a researcher would use this to combine more than one independent variable. This allows the research to test the limits of the effect of the IV and to test theories about the underlying mechanisms of behaviour.

Front 16

Why are factorial designs useful?

Back 16

- like two experiments in one

- economy of the subjects

- allow researchers to examine interactions

Front 17

When interpreting interactions from surveys what does is mean when the two lines are parallel? What if the midpoints of the two lines are different? Are the average values of the DV at each level on the IV on the x-axis different?

Back 17

- If the two lines are parallel there is an interaction present

- If the midpoints of the two lines are different than there is a main effect of B

- If the average values of the DV at each level on the IV are different there is a main effect of A

Front 18

If there is an interaction in your data set what should you do?

Back 18

- If the difference is significant, break down the results.

- Decide on one way to split the data but always report the full data set

Front 19

What are some types of complex designs?

Back 19

- IV x PV: manipulate one variable and cross that with a participant variable

- Designs with more than one IVs: for example, 3 factors with 2 levels each (2x2x2 Design)

- Designs with more than 2 levels of each IV: adding levels such as this allow us to determine if the relationship is linear

Front 20

How do quasi-experiments differ from true experiments?

Back 20

Quasi-experiments: have great external validity and tend to make weaker causal claims

True experiments: generally have greater internal validity and therefore able to make strong causal claims

Front 21

Compare and contrast the different types of special designs.

Back 21

1. Small n designs: these are single case designs

- Baseline designs: observe the behaviour before administering the treatment and then observe the behaviour after the treatment

- Reversal designs: observe the behaviour with and without treatment

2. Developmental designs:

- Longitudinal designs: observe the same group over time

- Cross-sectional designs: observe people at different ages at one particular point in time

- Sequential designs: observe people over different ages over time

3. Quasi-experimental designs:

Front 22

What are the five threats to the internal validity of quasi-experimental design?

Back 22

1. Maturation: participants change between the pretest and the posttest, but this change is not due to treatment

2. History: something happens in the real world at the same time as the onset of treatment & causes the effect

3. Testing: participants results in change due to being tested multiple times

4. Instrument decay: participants results change because the way they are measured or the way participants use the measure changes over time

5. Regression to the mean: extreme scores become less extreme on the second test

Front 23

What does regression to the mean look like?

Back 23

- If a variable is extreme on its first measurement, it will tend to be closer to the average on its second measurement and if it's extreme on the second measurement it will tend to have been closer to the average on it's first.

Front 24

Explain the different ways to visualize the data.

Back 24

- This depends on the type of variable. Discrete variables have no possible values between adjacent units on the scale. Continuous variable have an infinite number of values between adjacent units on the scale

- Discrete? Pie chart

- Continuous? Bar graph (has gaps between values) or histogram (no gaps between values)

Front 25

Describe the different ways to report the central tendency.

Back 25

Definition: summary of the central value of our data

- Mode: value that is observed most frequently. This is most often used to summarize discrete/categorical data.

- Median: This is the middle score when data is arranged in increasing order

- Mean: this is the arithmetic average

Front 26

Describe the different ways to measure the variability:

Back 26

Definition: A summary of the spread or variability of our data.

- Range: This is difference between the highest and the lowest scores

- Standard deviation: A quantity that is calculated to indicate the extent of deviation for a group as a whole

Front 27

Describe the different scales of measurement that exist.

Back 27

1. Nominal: these are general categories

2. Ordinal: there is a rank order, there is no equal distance between units

3. Interval: Equal distance between units but there is no true zero

4. Ratio: Equal distance between unites but there is a true zero

Front 28

What kind of scale should you use for the various scales of measurement that exist?

Back 28

Nominal/ordinal: percentage or mode

Interval/Ratio:

Are there outliers?

Yes - Median (or exclude them)

No - Mean (M) or standard deviation