REGRESSION
This is what we refer to in discussing a Y score when any and all X scores are equal to zero.
What is an intercept?
A sample size should like like this when calculating correlation.
What is N > 100, or a large sample size?
We use this tool to help us calculate these measures.
What is a cross-tabulation table?
When attempting to recode missing values of a variable, this is the first step.
What is "FREQUENCIES" - in order to gauge an understanding of missing data.
Correlation coefficients can be any value between these two values.
What are +1.00 and -1.00?
This is the equation for calculating the slope or regression coefficient.
What is Y = B + AX or Y = MX + B?
We must convert both X and Y to a Z scores before calculating this.
What is Pearson's R?
This measure demonstrates how exposure to a certain condition can influence the probability of experiencing the incdece.
What is relative risk (RR)?
We must employ this process before performing at data analysis.
What is data cleaning?
Instead of the NHST or the null hypothesis statistical testing, we use this measure to verify results.
What are confidence intervals?
We can add this type of information to our interpretation of a regression model to strengthen our discussion.
What is contextual information - information about the social/economic/policital context of the conditions of the model?
An assumption to meet when calculating correlation assumes this about the distribution shapes of X and Y.
What are fairly normal univariate and bivariate distribution shapes?
This measure tells us the how likely an incident would be in a group of exposed cases vs non-exposed cases.
What is relative risk (RR)?
When centering a variable on 0, we look at these two measures of central tendency.
What are the mean and the median?
The size of the P value tells us this about the value of D (Delta) - or the difference between the null hypothesis and the sample mean.
What is small P = big D; big P = small D?
This section of the regression output in SPSS tells us how different the value generated from the model is from the actual observed value, and can be positive or negative.
What is the standard error of the estimate or the residual?
The typical null hypothesis when calculating correlation is this.
What is no correlation or R = 0?
Degrees of freedom are dependent on this trait of a cross tabulation table.
What is the size - or 2x2, 2x3, etc.?
This is the first step in the string of steps when using the drop-down menu to run a T-Test to evaluate the null hypothesis.
What is "ANALYZE"?
This Excel command allows us to add together single cells of text into a string of text one cell.
What is "CONCATENATE"?
Using this trick allows us to assess the predictive usefulness of a X variable that is of primary interest while statistically controlling for more than one extraneous variable.
What is the use of multiple predictors?
R = +1.00 tells us this about correlation.
What is a perfect direct linear relationship between X and Y?
This measure provides an estimated decrease in a given incidence in a entire population if the risk of exposure is entirely eliminated by the group experiencing it.
What is population attributable risk (PAR)?
When running a T-Test, we check of this aspect in the drop down menu to get another statistical measure.
What is confidence intervals at 0.95?
A test of association is also called this.
What is a chi-squared test?