Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
96 Cards in this Set
- Front
- Back
True Results
|
- results that would be obtained by conducting a census of the entire population
- they are TRUE in the sense that they are free of sampling errors |
|
Chi-Square
|
- usual test of null hypothesis for differences between frequencies (number of cases or n)
- symbol is x2 - not a descriptive statistics |
|
Degrees of Freedom (df)
|
- calculates to see of null hypothesis is correct
- not a descriptive statistic - thought of as sub steps in mathematical procedure for obtaining the value of p Example: x2 = 4.00, df = 1, P< 0.05 |
|
P (Probability)
|
- when probability of null hypothesis is correct is .05 or less
- differences is STATISTICALLY SIGNIFICANT at the 0.05 level |
|
What happens if p is above 0.05?
|
Cannot reject null hypothesis & report it as statistically insignificant result
|
|
Univariate Analysis
|
- chi-square in which each participant is classified in only 1 way (ex. which candidate do they prefer)
|
|
When researchers study a sample, are the results called "true results" or "observed results"?
|
Observed results
|
|
Should the typical consumer try to interpret the value of df?
|
No
|
|
If a researcher fails to reject a null hypothesis, is the difference in question statistically significant?
|
No because the P value will be greater than 0.05
|
|
Bivariate Analysis
|
Analysis in which each participant is classified in terms of 2 variables in a chi-square test
|
|
Null Hypothesis
|
- No true difference
- if all members of the population had been studied, the research would've found no difference among the 3 methods 5 out of 100 times the null hypothesis is right |
|
Type I Error
|
When null hypothesis is true and it was rejected
|
|
Type II Error
|
null hypothesis should have been rejected but the significance test failed to lead the researchers to the correct decision
|
|
Frequency Distribution
|
- Describe quantitative data
- can construct a figure called a frequency polygon to show data |
|
Skewed
|
Have a tail on 1 side & not the other
|
|
Positive Skew
|
Skewed to the right
ex. Income |
|
Negative Skew
|
Skewed to the left
Ex. test basic math skills |
|
Mean
|
- most frequently used average
- balance point in a distribution - value around which the deviations sum to zero |
|
What is M a symbol of?
|
Mean of a population
|
|
What is m a symbol of?
|
Mean of a sample
|
|
What symbol do statisticians use?
|
x̄, pronounced "x bar"
|
|
Median
|
- the middle score
- used to describe the averaged of skewed distributions |
|
Mode
|
- most frequently occurring score
- seldom reported in formal research reports (b/c there can be more than 1 mode) |
|
Measures of Central Tendency
|
Synonym for "averages"
|
|
Standard Deviation
|
- one of 2 measurements used to describe variability
- symbol = S or SD for a population, s or sd for a sample - smaller variability = small standard of deviation |
|
Variability
|
Amount by which participants differ from each other
|
|
How many participants are within 1 standard-deviation unit of the mean if the distribution is normal?
|
68% of participants
|
|
Which average is usually reported when the standard of deviation is reported?
|
Mean
|
|
Range
|
- highest score minus the lowest score
- based on the 2 most extreme scores - considered unreliable |
|
Interquartile Range
|
- range between quarters (range of 2 middle quarters)
- range of the middle 50% of participants - used to help get around unreliability of using just range |
|
Should the mean or median be used w/ ordinal data?
|
Median
|
|
Large Interquartile Range = __________ (Lower/Higher) Variability
|
Higher
|
|
Pearson Product-Moment Correlation Coefficient
|
- most widely used coefficient for examining relationships between 2 quantitative sets of scores
- symbol = "r" called the "Pearson r" |
|
Coefficient of Determination
|
- Pearson r doesn't correlate to percetnages; convert Pearson r to this statistic to get percentages
- ex. if Pearson r = 0.50, you square 0.05 = 0.25 x 100% = 25% |
|
T test
|
- used to test null hypothesis regarding the observed differences between 2 means
|
|
When a researcher uses a large sample, are they "more likely" or "less likely" to reject the null hypothesis than when a researcher uses a small sample?
|
More likely
|
|
What does ANOVA stand for?
|
Analysis of Variance
|
|
If a researcher compares 2 means for significance, will ANOVA & the t test yield the same probability?
|
Yes
|
|
If ANOVA yields p > 0.05, are the differences statistically significant?
|
No
|
|
One-Way ANOVA (Single-Factor ANOVA)
|
When participants have been classified in only 1 way
|
|
If an overall ANOVA for 3 or more means is significant, it can be followed by what test to determine the significance of the differences among the individual pairs of means?
|
Multiple Comparisons Test
|
|
Is it possible for a small difference to be statistically significant?
|
Yes
|
|
What are the types of considerations for determining practical significance?
|
1) Cost-Benefit Analysis
2) Crucial Difference 3) Client Acceptability 4) Public & Political Acceptability 5) Ethical & Legal Implications |
|
Does a crucial difference need to be numerically large to be of practical significance?
|
Yes
|
|
Should the acceptability of a treatment to the clients be considered when determining practical significance?
|
Yes
|
|
True or False: Ethical considerations should play no role in the interpretation of the results of a study
|
False
|
|
For the results of 2 studies to be comparable, they need to be what?
|
Standardized (so both differences can be expressed on the same scale)
|
|
Effect-Size
|
- the magnitude (i.e. size) of a difference when it's expressed on standardized scale
|
|
What are the universally accepted standards for describing effect size?
|
There are none
|
|
What's the "effective range" of standard deviation units on both sides of the mean?
|
- effective range from 0.00 (no difference) to 3.00 (maximum difference)
|
|
What value of d is associated w/ "extremely large"?
|
d = 1.40+
|
|
When will a negative value of d be obtained?
|
When the control group's mean is higher than the experimental group's mean
|
|
Should a test of statistical significant be conducted before or after computing d?
|
After
|
|
What are 2 measures of effect size that are widely reported?
|
"d" & "effect size r"
|
|
Meta-Analysis
|
Set of statistical methods for combining the results of previous studies
|
|
What 2 types of random errors tend to be canceled out in the process of conducting a meta-analysis?
|
Random Sampling Errors
Random Errors of Measurement |
|
In a meta-analysis, what is the "main thrust of conclusions" based on?
|
Mathematical synthesis of statistical results of previous studies
|
|
What's the 2nd important characteristics of meta-analysis?
|
- systematic error from 1 researcher will be moderated by results from studies where those errors weren't made
|
|
What is 1 very important way various studies on a given topic differ?
|
Various researchers use different measures of the same variable
|
|
Cohen's d
|
- ranges from +3.00 to -3.00
- symbol of the measurement of effect size |
|
What should be done when the studies to be used in a meta-analysis have different sample sizes?
|
- d value should be figured for each study, averaged, & treated the same or weighted differently to take into account the varying number of participants
|
|
Is a meta-analysis more objective than a review consisting of narrative discussion?
|
Yes
|
|
Publication Bias
|
- final potential weakness of meta-analysis
- researcher may choose to not publish studies w/ insignificant results |
|
What's a solution for publication bias?
|
- researchers should search for studies listed in dissertations, convention papers, government reports, & other non-journal sources
|
|
Is purposive sampling widely used by qualitative or quantitative researchers?
|
Qualitative researchers
|
|
Purposive Sampling
|
- judgmental, selective, or subjective sampling
- type of non-probability sampling technique - units investigated are based off judgment of researcher |
|
Purposive Criterion Sampling
|
- purposive sampling when there are a number of criteria to be applied in selection of sample
|
|
When should participants from diverse sources be used?
|
In a qualitative study when the researcher has a broader interest than just a single source
|
|
Which type of research has more expensive data collection methods?
|
Qualitative
|
|
When has the data collection process become "saturated"?
|
- the point where several additional participants fail to respond w/ new info that leads to identification of additional themes
|
|
What does an interview protocol consist of?
|
Written directions for conducting the interview as well as a standard set of predetermined questions to be asked of all participants
|
|
Who should a pilot test of interview questions be conducted upon?
|
A few individuals that won't participate in study
|
|
Self-Disclosure
|
Process through which an interviewer can clear the air & clear the mind
|
|
Phenomenological Approach
|
Examining perceptions to acquire knowledge
|
|
How many participants are usually in a focus group?
|
6-12
|
|
What are the names of the individuals who lead focus groups?
|
Facilitator & Moderators
|
|
What's a "clear advantage" of using a focus group?
|
Reveals the evolution of perceptions in a social context
|
|
Nonparticipant Observation
|
Type of observation in which the researcher observes as an outsider
|
|
Ethnography
|
Field of research on cultural issues
|
|
Data Triangulation
|
- use of multiple sources for obtaining data on a research topic
- 2 or more types of participants are used |
|
Methods Triangulation
|
When a researcher interviews participants & then observes the behavior of the same participants
|
|
Researcher Triangulation
|
Type of triangulation that reduces the possibility that the results of qualitative research represent only the idiosyncratic views of an individual researcher
|
|
In peer review, what's a peer?
|
Person who provides the review (a.k.a. auditor)
|
|
Grounded Theory
|
- inductive method of analysis that can lead to theories of behavior
- starts w/ open coding |
|
Axial Coding
|
- 2nd step in grounded theory approach
- transcripts of data sources are reexamined w/ purpose of identifying relationships between categories & themes identified during open coding |
|
Constant Comparison
|
Technical term for constantly comparing each new element of data w/ all previous elements that have been coded
|
|
Consensual Qualitative Research (CQR)
|
- inductive process
- strives to have a team of researchers arrive at a consensus on the meaning of data collected |
|
What's the first step of CQR?
|
Code data into domains
|
|
What's the second step of CQR?
|
Writing short summaries of participants' ideas
|
|
What's the third step of CQR?
|
Cross-analysis of core ideas into categories based on similarities
|
|
How is an external stability check performed in a CQR?
|
By examining the data in addition to the interview transcripts
|
|
How is internal stability examined in CQR?
|
By determining the extent to which each category was general, typical, or variant
|
|
General Domain
|
Apply to all of participants
|
|
Typical Domain
|
Apply to half or more of participants
|
|
Variant Domain
|
Apply to less than half but more than 2 of the participants
|
|
CQR method requires the use of a(n) _________
|
Auditor (outside expert
|