ch 11 understanding statistics in research Flashcards
type of stats: saying what data shows
-used to present quantitative data
-helps make sense of a large amount of data
(ex: GPA)
descriptive stats
3 characteristics of a single variable (in descriptive stats)
-distribution
-central tendency
-dispersion
The summary of the frequency of individual values or ranges of values for a variable, can be expressed as percentages or group
distribution
what is included in central tendency
-mean (average)
-median
-mode (most frequent)
what is included in dispersion
-range
-standard deviation
Trying to reach conclusions that extend beyond the immediate data alone
inferential stats
represents the probability of an outcome as it relates to random chance;
how likely it is that a treatment group is significantly different from a control group
p-value (0-1)
p-value cutoff point for showing real-
world significance
0.05
inferential statistical tests for differences
-chi square
-t-test
-analysis of variance (ANOVA)
type of inferential stat test used for differences:
-used with nominal
-compares observed frequencies to expected frequencies
chi square
type of inferential stat test used for differences:
-assesses whether the means of two groups
are statistically different from each other
-Appropriate when you want to compare the means of two groups
(ex: systolic BP before and after Tx)
t-test
type of inferential stat test used for differences:
-tests the significance of group differences between two or more groups
-only determines that there is a difference
between groups, but doesn’t tell which is different
ANOVA (analysis of variance)
inferential stat tests for relationship
-correlation
-multiple regression
inferential stat test for relationship:
-used with two variables to determine a
relationship/association
-does not distinguish between independent and dependent variables
correlation
inferential stat test for relationship:
-used with several independent variables and one dependent variable and used for PREDICTION
-identifies the best set of predictor variables
(ex: drug use, alcohol use, child abuse, suicidal tendencies)
multiple regression