Statistical Analysis and Results Section Flashcards
Results section should contain
Narrative description of statistical outcomes
Tables and figures that summarize findings
Statements of support of the hypotheses or rejection of the hypotheses
Interval or ratio measurement data may be described in 4 ways:
Central tendency
Variability
Skewness
Kurtosis
Average score of a group
Central tendency
3 measures of central tendency
Mean, median, mode
How much the scores vary from the average
Variability
3 measures of variability
Range, variance, standard deviation
The lack of symmetry of the distribution of scores
Skewness
the general shape of the distribution of scores
Kurtosis
When does normal distribution happen
when the middle scores occur most often and the lower and higher scores do not occur often.
If the data is not normally distributed then….
nonparametric statistical procedures are used
more powerful than nonparametric statistical procedures.
Parametric statistics
4 Parametric statistical procedures:
Normal distribution of the data
Interval or ratio level of measurement
If 2 or more data distributions are analyzed, their variances should be similar
Large sample size
When are Nonparametric statistics used
when one or more of these are not met
Statistical significance testing involves
testing the null hypothesis in the context of the data.
likelihood that one event will occur, given all the possible outcomes
Probability
probability of the findings
P value
p < .05 means
NULL HYPOTHESIS IS REJECTED
p > .05 means
NULL HYPOTHESIS IS NOT REJECTED
can be correlational or inferential
Data analysis
level of significance
.05 or less (p < .05)
evaluate relationships among data
Correlational statistics
often described using correlation coefficients
RELATIONSHIPS W/ IN DATA
A perfect positive relationship between two variables is indicated by
1.0
A perfect negative relationship is indicated by
-1.0
The absence of a relationship is indicated by
ZERO
A small number indicates a
weak relationship between two variables
A large number indicates a
strong relationship between two variables
The square of the correlation coefficient is used to assess
PRACTICAL MEANING
Variables that are correlated can be described as
varying together, but there may be no cause-effect relationship
Presenting the results of correlational statistics involves 4 types of analysis/tables
Regression analysis
Bivariate analysis
Multivariate analysis
Contingency table
Regression analysis
Bivariate analysis
Multivariate analysis
Chi square
Contingency table
nonparametric test applied to nominal data, comparing observed frequencies within categories to frequencies expected by chance
Chi-square
evaluate differences among data, either between-subjects or within-subjects
Inferential statistics
used to compare two different groups
Independent t-test
used for within group comparisons
Dependent t-test
simultaneous comparison of several means
ANOVA- analysis of variance
only one independent variable
One-way ANOVA
two independent variables
Two-way ANOVA
between subjects comparison (ordinal data)
Kruskal-Wallis one-way ANOVA
from related samples (nominal data)
Cochran Q test
within-subjects comparison (ordinal data)
Friedman two-way ANOVA
from independent samples (nominal data)
Chi-square
multivariate analysis of variance
MANOVA
analysis of covariance
ANCOVA
Bonferroni correction
Multiple t-test
Effect size is
a quantitative measure of the difference between two groups
Cohen’s d - effect size estimator used to
compare the means of two or more groups
