1/44
Looks like no tags are added yet.
APA high-level style
second level
includes guidelines for clear expression of ideas
emphasis on avoiding biased language
2 themes
APA-style writing is formal
straightforward writing
APA low-level style
third level
guidelines for spelling, grammar, references, citations, statistics, and numbers
empirical research report
article that presents the results of one or more research study
review articles and theoretical articles
final manuscripts
dissertations, theses, and other student papers
professional conferences
events for researchers to present their research to each other
oral presentations
poster session (has become increasingly popular)
descriptive stats
a set of techniques for summarizing and displaying data
distribution
every variable has one. it’s how the scores are distributed across the levels of that variable.
frequency tables (e.g. histograms) display them
negatively skewed

positively skewed

central tendency
the middle point where the scores tend to cluster
unimodal and symmetrical
mean, median, and mode
will all be close to each other at the peak of distribution

bimodal/asymmetrical distribution
mean, median, and mode are very different

variability
how much the scores vary around their central tendency. range measures this
standard deviation
average of scores and mean
percentile ranks and z-scores
used to determine where an individual score is in a distribution
effect size
used to determine the strength of a statistical relationship
cohen’s d
used to measure the effect size for differences between any group or condition
when can pearson’s r be misleading?
the relationship is nonlinear
one or both variables have a limited/restricted range in the sample
error bars
represent the variability in each group/conditions
sampling error
the random variability in a statistic across samples
null hypothesis testing (NHST)
deciding between two interpretations of a statistical relationship (H0)
when do you reject the null hypothesis?
if the sample relationship would be extremely unlikely in the population
when do you retain the null hypothesis?
if the sample relationship would be extremely likely in the population
weak relationship based on medium or larger samples are…
never statistically significant
strong relationship based on medium or larger samples are…
always statistically significant
pracitical significance
the importance or usefulness of a result in real-world context
p-value
probability of finding a sample or more extreme value would be unlikely if the null were true
one sample t-test
compares sample mean to hypothetical population mean
when do we reject the null in a two-tailed t-test?
if the t-score is extreme in either direction
dependent samples t-test
used to compare two means for the same sample tested at all different times or under two different conditions
independent samples t-test
compares the means of 2 separate samples
M1 and M2
analysis of variance (ANOVA)
used when there’s two groups or conditions to be compared
one-way ANOVA
used for between-subject design with a single independent variable
mean-squares between groups (MSp)
one estimate of the population variance
mean squares within groups (MSw)
also an estimate of the population variance, but it’s based on the difference among the scores within each group
part hoc comparisons
selected pairs of group means to determine which ones are different from others
repeated-measures ANOVA
multiple measures of the same variable taken on the same or matched subjects in different conditions or over two or more time periods
factorial ANOVA
used when more than one independent variable is included in a factorial design
type I error
false positive
alpha
type II error
false negative
beta
file drawer problem
when researchers obtain non-significant results, they tend to not submit them for publications. if they do submit them, journal editors and reviewers don’t accept them
p-hacking
intentionally making decisions to increase the chances of having statistically significant results
makes a lot of type I errors
bayesian stats
the researcher specifies the probability that the null hypothesis and any important alternative hypotheses are true before the study
replicability crisis
the inability of researchers to replicate earlier research findings
criticism about null hypothesis testing
many researchers misunderstand it. some suggest to use confidence intervals instead.