Q1 behavioral variability Flashcards
variability
the degree to which scores in a data set vary or differ from one another
how do we distinguish between individual differences and behavioral variability? what factors account for or explain variability?
research design, measurement of variable, and statistical analyses
how does research design help us parse variability?
we can identify the source of variability using experiments or non-experimental designs (correlational, longitudinal, cross-sectional)
how does the measurement of a variable help us parse variability?
to correctly parse variability the measurement of a variable must be appropriate and quantifiable
how do statistical analyses help us parse variability?
they summarize and answer questions about observed variability
descriptive stats give us averages, percentages, and frequencies which summarize and simplify our data
inferential statistics allow us to make assumptions or generalizations to a wider population; they also allow us to directly test hypotheses (is the relationship meaningful or significant?)
what are measures of central tendency? (definition)
they help us to see where the general distribution lies in our data set
what are the three measures of central tendency?
- mean
- median
- mode
what information do measures of variability give us?
they tell us how loosely or tightly packed scores are around the mean
what are 3 different ways we can measure variability? (2/3 are kinda the same)
- range
- standard deviation
- variance
range
max - min values (but we miss out on the middle values)
variance and standard deviation
subtract every score from the mean and square each value to get rid of negative values; sum of squares then divide by N - 1 (and take the square root to get SD)
what are the two types of variance?
- systematic
- error
systematic variance
variance that occurs in a predictable and orderly fashion; aka your variable of interest
how much variability in y is systematically explained by variable x?
error variance
variability not accounted for by the explanatory variable
variability in y that is unrelated to variable x
this comes from factors that we fail to measure in the study
what is the problem with having high error variance?
we decrease our understanding of the full picture or the systematic variance that we are looking for (more noise/static); makes it harder to detect a true effect
how do we parse out variance to reveal systematicity?
using regression, t-tests, ANOVA, etc.
what do p values tell us?
whether there is a significant difference or relation between two variables or groups
what do effect sizes tell us?
effect sizes measure the magnitude or strength of an observed difference or the strength of the association between variables of interest
is the difference between groups or the relation between two variables impactful or meaningful?
do effect sizes help us decide whether to reject the null hypothesis?
no
R^2
used for correlational/regression designs
if R^2 = 0.2, then 20% of variance in variable y is explained by variable x (80% is unaccounted for)
n^2
proportion of variance, ANOVA
cohen’s d
mean differences, t-tests
which effect size is used in correlational studies?
R^2, pearson’s correlation coefficient
which effect size is used in tests of the mean difference between two groups (t-tests)?
cohen’s d
