Final Flashcards
Describes variables and relationships between variables (table 1)
Descriptive statistics
Used to determine if an IV has a significant effect on a DV
inferential statistics
we use this/these measures of central tendency when analyzing ratio level data
mean, median, mode
We use this/these measures of central tendency when analyzing ordinal level data
median, mode
We use this/these measures of central tendency when analyzing nominal level data
mode
What are the 5 ways to report data?
- %
- % change
- rates
- ratios
- proportions
batting average is an example of reporting data as a:
proportion
(# of hits vs. # of a bats)
If there are 25 nurses now, compared to 17 a few years ago, we are looking at the:
% change
If looking at divorce, suicide, crime… we look at:
(# of incidents/total population) x 100
This is an example of:
rates
if looking at something out of 100,000 people, this is an example of:
ratios
these are always between 0 and 1
proportions
1 SD from the mean
68.2%
2 SD from the mean
96%
Represents a comparison of one thing to another
ratios
(new-original / original) x 100
percentage change
represents the frequency of something for a standard sized unit
rates
also called Gaussian distribution
Normal distribution
Represents the distance above or below the mean, in standard deviation units, of any raw value in a distribution
z score
ranges from -3 to +3
z score
What are 3 ways to describe the relationship between variables?
- crosstab analysis
- comparison of means
- correlations
In order to do a chi square, the DV must be ____
nominal
if you want to figure out what gender of people by certain cars, you would do this type of stat design
chi-square
when studies that find a significant difference are easier to publish, this is called
publication bias
Differences among groups for a single IV that are significant, temporarily ignoring all other IVs
main effect
differences among groups of a single IV that are predictable only by knowing the level of another IV
interaction effect
When you get a false positive, this is called:
type 1 error
the degree of risk you are willing to take to reject the null hypothesis when it is true
type 1 error
