Chapter 4—Sampling, Measurement & Hypothesis Testing Flashcards
(24 cards)
How do we evaluate a measure’s proficiency?
Validity and Reliability are tested
What are the four types of measurement scales?
- Nominal scales: categorical
- # of tea drinkers vs coffee drinkers by age
- Ordinal scales: ordered categories where the intervals between each category isn’t necessarily equal
- very unhappy, unhappy, neutral, happy, very happy
- Interval scales: Equal intervals between values but no true “zero” point; zero doesn’t mean the absence of what’s being measured.
- 0 degrees Celsius doesn’t mean there’s no temperature
- Ratio scales: quantitative, 0=none
- 0 kills in fn means the absence of kills
What is reliability? What does it tell us?
Is the measure consistent and repeatable; measures high in reliability contain a minimum amount of measurement error
Reliability tells us if a particular score on a measure means anything. If someone gets different results every time…
Define Measurement Error
An error produced by something that creates inaccuracies in measurement. Can be introduced by participants (sleepiness) or the tool itself (poorly calibrated)
What is validity?
The extent to which a measure of X truly measures X and not Y
What are the 7 types of validity discussed in this chapter?
Content validity: do the items on a test comprehensively measure the entirety of X?
Face validity: does the measure seem valid to those taking it
- alerts us to the experimental realism
Criterion validity: is the measure related to a real world outcome?
- Predictive validity: Does the measure predict future
behaviour or performance?
- Concurrent validity: Is the measure correlated with an
existing, established measure taken at the same time?
Construct validity: Does the test truly measure the theoretical construct it’s intended to measure
- Convergent Validity: does a test correlate with other
tests that measure similar constructs.
- Discriminant Validity: does a test not correlate with
tests that measure unrelated constructs
What is a sample?
The people participating in the study; a subset of the population that should reflect the population parameters
What is Probability Sampling?
each individual of the population has a certain probability of being selected for the sample
What is random sampling?
Each member has an equal chance of being selected
What is stratified sampling?
A random sample except important subgroups are proportionately represented
What is cluster sampling?
Randomly selecting clusters of “people that have some feature in common”, and tests all people within those selected clusters
What is nonprobability sampling?
A sampling technique where not all individuals in a population have an equal chance of being selected
What is a convenience sample?
People who meet the general requirements of the study and are recruited in a variety of non-random ways
Describe purposive sampling
A researcher targets a particular group of individuals
Describe Quota Sampling
A researcher mimics stratified sampling but does so non-randomly. They recruit until they meet the quota instead of randomly selecting
What is snowball sampling?
One someone has been surveyed, the researcher asks them to help recruit more people via friend networks.
What are descriptive statistics?
stats used to summarize and present the data collected
What are inferential statistics?
Stats that allow you to draw conclusions about your data that can be applied to a wider population (ex. Ho testing)
What does the alpha (a) level = 0.05 signify?
Essentially means that, if your sample would reject the null hypothesis at a=.05, there is a <5% chance your supporting evidence occurred by chance.
What is a p-value?
A value that signifies the significance of your findings.
Significant if p < a
What is an effect size?
The amount of influence one variable has on another; the amount of variance in DV that can be attributed to the IV
Provides a standardized way to compare results across different studies
What is a meta-analysis?
Uses Effect Size analysis to combine the results of many experiments that use the same variables, even if these variables may have different operational definitions.
Helps determine if general patterns occur in the data.
What is the 95% confidence interval?
A range of values that is expected to include the population parameter with 95% accuracy.
Non-overlapping CIs indicate a meaningful difference between 2 conditions of study
What signifies a significant difference between two groups in Null Hypothesis testing?
A noticeable difference between the mean scores and a relatively small amount of variability within each group
