Statistics

Question 1

regression

Answer 1

– all variables examined are continuous  Linear regression – degree of dependence between one variable and another. Data is on scatter plot, one-way influence of one variable on another.

Question 2

correlation

Answer 2

• all variables examined are continuous. Unlike regression makes no assumptions about which variable is influencing the other.  If correlation coefficient is 1, perfect. If -1, opposite. 0, random.

Question 3

chi squared

Answer 3

when all variables are categorical, looks at if 2 distributions of categorical data differ from each other.  Null hypothesis vs. alternative hypothesis.

Question 4

t test

Answer 4

compares mean values of a continuous variable (dependent) between 2 categories/groups, ex. comparing mean of a group to a specific value. Can also compare means of 2 groups.  Two-tailed = possibility of relationship in both directions, one-tailed = one direction

Question 5

anova

Answer 5

similar to t-test, compare distributions of continuous variable between groups of categorical variable, but can be used for 3+ groups. If value doubles, 100% increase

Question 6

cross sectional

Answer 6

look at a group of different people at one moment in time

Question 7

cohort

Answer 7

following a subset of population over a lifetime. A cohort is a group of people who share a common characteristic (ex. people born and exposed to same pollutant/drug/etc.) in period of time. - Opposite of case-control… cohort follows two groups into the future

Question 8

longitudinal

Answer 8

data is gathered for the same subjects repeatedly over a period of time, can take years or decades.

Question 9

case control study

Answer 9

observational study where 2 groups differing in outcome are identified and compared to find a causal factor. Ex. comparing people with the disease with those who don’t but are otherwise similar.  is a retrospective look back!

Question 10

clinical trial

Answer 10

highly controlled interventional studies

Question 11

randomized controlled trial

Answer 11

– people studied randomly given one of treatments under study, used to test efficacy/side effects of medical interventions like drugs. Gold standard for a clinical trial.

Question 12

internal validity

Answer 12

extent to which a causal conclusion based on a study is warranted. Confounding factors often impact the internal validity of an experiment.  Extent to which a piece of evidence supports a claim about cause and effect • = the strength of assigning causes to outcomes  BASICALLY… Is the research sound and did it avoid confounding variables??

Question 13

external validity

Answer 13

Whether results of the study can be generalized to other situations and other people. To protect external validity, sample must be completely random, and all situational variables must be tightly controlled.

Question 14

construct validity

Answer 14

whether a tool is measuring what it is intended to measure

Question 15

confounding variables

Answer 15

– changes in dependent variable may be due to existence of or variations in a third variable  Affects dependent and independent variables, causing a spurious association  Correlates and explains both the dependent/independent variables.

Question 16

mediating variables

Answer 16

provides an explanation for why two things are related

-it mediates the IV/DV

Question 17

moderating variabels

Answer 17

affects the strength of the relationship between 2 variables

Question 18

reliability

Answer 18

RELIABILITY = CONSISTENCY Test-retest reliability: Test it again and it’s the same Inter-rater reliability: Two different people measure the same thang

Question 19

deductive v inductive reasoning

Answer 19

Deductive reasoning = start with a general principle and apply it to a special case

• Like top-down

Inductive reasoning = start with a special case and apply it to a general principle

Like bottom-up