Statistics Flashcards
Bell Curve
Normal distributions with mean of 100. 34.1% to right of middle and 34.1% to left of middle average
Mean
Average
Mode
that occurs the most
Median
Middle #
Negative skew
tail is on the left
Positive skew
tail is on the right
What is central tendency
tendency of scores to cluster around some central value
If test scores are positively skewed how would one create a better symmetrical distribution
by deleting tough questions and substituting them with easier ones
Standard Error of Measurement
how much difference to expect from one sample to another (same as confidence interval) repeated measures on same instrument distributed around true score
Standard Deviation
deviation from the mean. SD of 100 it is 15 points above or below for 1 SD
4 ways for test interpretaion
Standard Score-mean of 100; Scaled Score-raw score translated into comparable score, Mean of 10 SD of 3, T score, Percentile Rank
What is a T-score
normal distribution-likelihood of observations-how does data compare to what is expected. Mean of 50 and SD =10
Z score
how many SD a data point is from the mean in a normal distribution. ranges from -3 SD to +3 SD-way to compare results to a normal population
Effect Size
measurement of absolute magnitude. helps determine if the difference is real or due to change factors. strength of relationships between two variables
Dependent Variable
variable being observed and measured-event expected to change when independent variable is manipulated
Independent Variable
variable that is changed or manipulated
Type 1 error
said there was an effect by wasn’t one. rejecting null hypothesis when it was actually true (False Positive)
Type 2 error
There was an effect and you missed it-failing to reject the null hypothesis (False Negative)
Null hypothesis
there is no significant difference
Pearson Correlation Coefficient (r)
measure correlation varies b/w -1 and +1 effect size . effect size is low if value of r is around .1, medium if r varies around .3 and large if r varies more than .5
Regression
used to measure predictions-determine the strength and character of relationships between one dependent variable (Y) and a series of other independent variable
Cohen’s D
Used to measure effect size when you’re comparing a treatment to a control group
internal reliability
measure of how consistently the items in a test measure the same concept
external reliability
measure of a test to produce the same results over time and across different individuals.
