Exam 2 Flashcards
Mean
average
Median
mindpoint
Mode
the most frequent
Range
Difference between highest and lowest values
Variance
Average of squared deviations from the mean
Standard Deviation
- Square root of variance
- describes the variability of a single sample
Frequency Distribution
Number of times that each value appears
Often converted to %
Ex. What is your gender
- Level of measurement
- Central Tendency
- Variability
Level of measurement: Nominal
Central Tendency: Mode
Variability: Frequency
Ex. Rank these five brands in order of your preference
- Level of measurement
- Central Tendency
- Variability
Level of measurement: Ordinal
Central Tendency: median, mode
Variability: frequency
Ex. On a scale of 1 to 5 how does Starbucks rate on the variety of its drinks?
- Level of measurement
- Central Tendency
- Variability
Level of measurement: scale
Central Tendency: mean, median, mode
Variability: frequency, SD, range
Ex. About how many times did you buy fast food for lunch last week?
- Level of measurement
- Central Tendency
- Variability
Level of measurement: scale
Central Tendency: mean, median, mode
Variability: SD, range
Descriptive Statistics
are computed from information provided by a sample
Statistical inference
uses sample statistics along with sample size to make inferences about a population
Central-limit Theorem
- As sample size increases, distribution of sample means, randomly selected, approaches normal distribution
- If assume approximately normal distribution, can make inferences about a variable
Standard error
represents the standard deviation of a theoretical sampling distribution of means
How do we tell if a result is statistically significant?
How many standard errors away from the comparison value is your result?
- If more than two standard errors away from the comparison value, conclude the difference you found is unlikely to have occurred by chance and is therefore significantly different from the comparison value.
Comparison value
is that difference in the population parameters is equal to zero—meaning there is NO DIFFERENCE between groups
How Do You Know When the Results Are Significant?
If the result falls outside of 2 standard errors from 0, it is not likely that the true difference is 0. Rather, it is likely that there is a real statistical difference between the two means in the population.
In other words, there is such a large difference between the means that it probably didn’t occur due to chance (sampling error) and it is likely the result of a true difference in population means.
ANOVA
Differences between three or more means from independent samples
Paired Samples t test
Differences between means when there are two measures for each respondent like a brand rating pre/post viewing an ad
Covariation
amount of change in one variable systematically associated with a change in another variable
Correlation coefficient
- between the range of −1.0 and +1.0
- communicates strength and direction of linear relationship between two scale variables
size indicates strength of association - sign (+ or −) indicates the direction of association
Coefficient Range:
+.81 to +1.0; -.81 to -1.0
Strong
Coefficient Range:
+.61 to +.80; -.61 to -.80
Moderate
Coefficient Range:
+.41 to +.60; -.41 to -.60
Weak
Coefficient Range:
+.21 to +.40; -.21 to -.40
Very Weak
Coefficient Range:
+.20 to -.20
None
correlation analysis requires….
two scale level variables
Regression analysis
is a predictive analysis technique in which one or more variables are used to predict the level of another by use of the straight-line formula
Bivariate regression
means only two variables are being analyzed, and researchers sometimes refer to this case as “simple regression.”
Independent variable
used to predict the dependent variable (x in the regression equation)
Dependent variable
that which is predicted (y in the regression equation) (text calls it the “criterion variable”)
Y = a + bx
What does each component mean?
Y is dependent
A is intercept
B is slope
X is independent
Dummy variable
nominal variable that can be shown as either a 0 or 1 (ie if January then 1 otherwise 0)
R-square
(aka coefficient of determination), is a measure of the strength of the linear relationship in multiple regression
- Indicates how well the independent variables can predict the dependent variable.
Independence assumption
the independent variables must be statistically independent and uncorrelated with one another (the presence of strong correlations among independent variables is called multicollinearity).