Week 4: Developing Your Research Proposal Flashcards
The entire group of individuals relevant to your research
Population
A subset of individuals drawn from the population of interest
Sample
A sample that shares the essential characteristics of the population from which it was drawn
Representative sample
A method in which every member of a given population has an equal chance of being selected into a sample
Random Sampling
Sampling methods for which the probability of a person being selected into a sample is known
Probability sampling
A technique whereby a population is divided into homogeneous group, called strata, along some key dimension (e.g., race/ethnicity), and then random samples are drawn from within each of the strata.
Stratified random sampling
The intentional over-recruitment of underrepresented groups into a sample to ensure there will be enough representation of those groups to make valid research conclusions
Oversampling
A sample in which members of the population are not all given an equal chance of being selected
Non-probability sample
A method of sampling that makes use of the most readily available group of participants
Convenience sampling
A method of sampling in which participants are asked to help recruit additional participants
Snowball sampling
An instance where participants electively place themselves into a particular sample (or they opt out of participation)
Self-selection
The difference between the actual or true value of what you are measuring and the result obtained using the measurement instrument
Measurement error
Responses are unordered categories. Although responses might be assigned numbers for coding purposes (e.g., Liberal = 1, Conservative = 2), the numerical assignments are arbitrary
Nominal
Responses are ordered categories (“greater than” or “less than” relationships make sense)
Ordinal
Responses are numerical and the differences between points on the scale are numerically meaningful
Interval
Responses are interval and there is a meaningful 0 value.
Ratio
A statistical test that makes few assumptions about the population distribution and may be applied to nominal and ordinal measurements
Non-parametric test
A statistical test that requires that the measurement scale of your data be interval or ratio and makes strong assumptions about the distribution of measurements in your population
Parametric test
Items that ask participants to rate their attitudes or behaviour using a predetermined set of responses that are quantified. An example would be a 5-point rating scale that asks about level of agreement from 1 (strongly disagree) to 5 (strongly agree)
Likert-type ratings
The probability that your study will be able to detect an effect in your research, if such an effect exists
Statistical power
A series of computations that help you determine the number of participants you will need to successfully detect an effect in your research
Prospective power analysis
The magnitude of an effect (such as a difference of means or a relationship between two-variables)
Effect size
A series of computations that help you determined how much power you had in a study after the fact
Retrospective power analysis
An indicator of the probability of obtaining an effect size as large as (or larger than) the one you obtained; also indicates that the differences between groups are not due to sampling error
Statistical significance
The process of deliberately manipulating factors in your research to maximize your chance of uncovering a statistically significant finding (p<0.5)
p-Hacking
The art of juggling choices
1) Participant Recruitment Issues
2) Time Constraints
3) Money Constraints
4) Equipment Constraints
5) Make the Best Choices
