Statistics

Question 1

A symmetric distribution with equal mean, median, and mode
Standard deviations follow the 68 - 95 - 99.7 rule

Answer 1

Normal distribution

Question 2

A normal distribution with a mean of 0 and a standard deviation of 1

Answer 2

Standard distribution

Question 3

Asymmetric distribution. Named either left-skewed or right-skewed by what side the tail is on

For right skew distributions, mean > median.
For left skew distributions, median > mean.

Answer 3

Skewed distribution

Question 4

IQR = (third quartile) - (first quartile)

Answer 4

Interquartile range

Question 5

Line inside the box represents the median

Upper and lower quartiles (75%, 25% of data rely beyond, respectively) indicated by upper and lower edges of the box

Maximum and minimum datapoints represented by the whiskers
95% confidence interval indicated by the notches. If notches between boxes do not overlap → statistical significance indicated

Answer 5

Box-and-whisker plot

Question 6

The factor that is being changed or manipulated

Answer 6

Independent variable

Question 7

The outcome that is being measured

Answer 7

Dependent variable

Question 8

Variable that influences both the independent and dependent variables. Confounds the ability to determine causality
Example: Tobacco smoking confounds the relationship between chewing tobacco and mortality

Answer 8

Confounding variable

Question 9

Variable that influences the strength of a relationship between two other variables
Example: Amount smoked influences the relationship between cigarette smoking and mortality

Answer 9

Moderating variable

Question 10

Variable that explains the relationship between two other variables

Example: Increased cancer risk helps explain relationship between smoking and mortality

Answer 10

Mediating variable

Question 11

The r value, describing the linear relationship between two variables
Ranges from -1 to 1 and describes the direction and strength of an association
Example: -0.30 is a negative and moderate association, whereas +0.90 is a positive and strong association

Answer 11

Correlation coefficient

Question 12

Study variables that are not directly measurable (e.g. depression, happiness) are defined in a way that they can be measured

Answer 12

Variable operationalization

Question 13

p value definition

Answer 13

Describes the likelihood of finding a difference when, in reality, there is no difference between the groups (null hypothesis is true)
Statistical significance threshold for the study is most often set at α = 0.05 (5%)
This means if p < 0.05, then we reject the null hypothesis; if p > 0.05 then we fail to reject the null hypothesis

Question 14

Requires categorical variables
Compares null hypothesis vs alternative hypothesis, looking to see if the two distributions of categorical data differ

Answer 14

Chi-square

Question 15

Requires continuous variables
Compares the mean values of continuous variables of 2 groups

Answer 15

T-test

Question 16

Requires continuous variables
Similar to t-test, but can be used for 3 or more groups

Answer 16

ANOVA

Question 17

A threshold that describes the chance that results of a study are due to random chance rather than causal effect, usually set at α = 0.05

Answer 17

Statistical significance

Question 18

When the null hypothesis is incorrectly rejected, in other words a false positive

Answer 18

type 1 error

Question 19

When the null hypothesis is incorrectly supported, in other words a false negative

Answer 19

type 2 error

Question 20

States that no significant difference or relationship exists between study variables

Answer 20

Null hypothesis

Question 21

factors that can cause a result of a study to differ from the true result

Answer 21

bias

Question 22

Bias introduced by the selection process of including subjects in a study (e.g. study population is not representative of the whole population)

Answer 22

selection bias

Question 23

Study subjects behave differently when they know they’re being studied

Answer 23

Hawthorne effect

Question 24

Bias in survey studies where people answer in a way considered socially desirable and acceptable

Answer 24

Social desirability bias

Question 25

Administration of an inactive substance or sham procedure corresponds to improved symptoms

Answer 25

Placebo effect

Often related to a person’s belief that a treatment will work

Comparison between placebo group and treatment group is used to determine the true benefit of a given treatment (where the patients are unaware if they received the treatment or placebo - known as blinding)

Question 26

Certain individuals in a population have a greater chance of being selected for a study than other individuals, resulting in a sample that does not accurately reflect the population

Answer 26

Sampling bias

Question 27

Tendency of a person to answer questions on a survey untruthfully or misleadingly

Answer 27

Response bias