Stats Midterm

Question 1

Scientific Method

Answer 1

Process of systematically collecting and evaluating evidence to test ideas and answer questions.

Question 2

Steps of Scientific Method

1

Answer 2

Come up with a research question that is specific/testable and review the literature on that topic up to this current date

Question 3

Steps of the Scientific Method

2

Answer 3

Design a study by considering the target and sample populations, comparison groups, whether the study would be observational/experimental or prospective/retrospective, and consider confounding factors

Question 4

Steps of the Scientific Method (3)

Answer 4

Execute the study and collect the data

Question 5

Steps of the Scientific Method

4

Answer 5

Look at the findings: Descriptive statistics

-use suitable graphical displays as necessary

Question 6

Steps of Scientific Method

5

Answer 6

Test the hypothesis: Significance tests

-estimate the magnitude of any statistical associations

Question 7

Variables in Studies

Answer 7

Predictor (independent)

Outcome (dependent)

Question 8

Types of Variables

Categorical (Qualitative Data)

Answer 8

Binary - Sex (male, female)
Nominal - Race (White, Black, Asian, etc.)
Ordinal - Grade in school (think order 1st, 2nd, etc)

Question 9

Types of Variables

Continuous (Quantitative Data)

Answer 9

Interval - Kinsey scale, GPA

Ratio - Age (has an absolute 0)

Question 10

Hypothesis vs Null Hypothesis

Answer 10

Hypothesis: Statement in hypothesis testing about the predicted relation between populations (often a prediction of a difference between population means)

Null Hypothesis: Statement about the relation between populations that is the exact opposite of the research hypothesis. (there is no difference)

Question 11

Measures of Central Tendency

Answer 11

Mean - sum of all the scores divided by the number of scores (M = ΣX / N)

Mode - Most frequently occurring number in a distribution

Median - The value at which 1/2 the ordered scores fall able and 1/2 of the scores below

Question 12

Measures of Central Tendency cont.

Answer 12

The mean, median, and mode coincide in a normal distribution

In a skewed distribution, the mean is “pulled” to toward the tail of the distribution

Mean is the most stable measure of central tendency

Question 13

Z-Score

Answer 13

The number of standard deviation a score is above or below the mean
Z = (X - M)/S
X - Single value
M - Sample mean
S - Sample SD

X = (Z)(SD) + M

Sum of a set of z-scores is always 0 because the mean has been subtracted from each score

The SD of a set of standardized scores is always 1 because the deviation scores have been divided by the standard deviation

Question 14

Normal Curve

Answer 14

is symmetrical and unimodal with most scores falling near center and few at the extremes

Question 15

Frequency

Distributions

Answer 15

Unimodal: Having one mode
Bimodal: Having two modes
Multimodal: two or more modes
Rectangular: all values have the same frequency
Skewed: scores pile up on one side of the middle

Question 16

Kurtosis (just in case)

Answer 16

The extent to which a frequency distribution deviates from a normal curve in terms of if the curve is more peaked or flat than a normal curve.

More peaked: “heavy-tailed”
More flat: “light-tailed”

Question 17

Type I vs Type II Errors

Answer 17

Type I: a false positive. When results are concluded to be statistically significant when in reality there is no effect. Rejecting the null when it should have been accepted.

Type II: missed opportunity. Concluding results are not significant when they really are. Accepting the null when it should have been rejected.

Question 18

Central Limit Theorem

Answer 18

If taking population with a mean of μ and a standard deviation of σ and take a relatively large random sample will result in a normal distribution

Question 19

Standard Deviation

Answer 19

The most common way of describing how spread out a group of scores is from the mean. The square root of the variance.

Question 20

Alpha Levels

Answer 20

A p-value that tells you the probability of rejecting the null hypothesis when the null hypothesis is true (Type I Error)

Question 21

Effect Sizes

Answer 21

A measure of the difference between population means. Also, how much something changes after a specific intervention.

Question 22

P Value

Answer 22

P-value is what you have chosen to be the significance level
P of .05 means there is a 5% chance that we have made an error
P-value is what you get from your output

Question 23

Z Tests

Answer 23

A hypothesis test used when a single sample and the population variance is known. Knowing both the population mean and the population variance is quite rare

Question 24

T Tests

Answer 24

Single sample t-test: testing a sample mean with a population mean

Independent sample t-test: testing the difference between two sample means

Paired sample t-test: (dependent samples t-test) two scores for each person (before and after) within the same sample.

Question 25

Degrees of Freedom (df)

Answer 25

N - 1 (one sample t test) | N1 + N2 - 2 (independent samples t test)

Question 26

T distributuion

Answer 26

The similar distribution to a normal distribution. Will become closer to a normal distribution with a larger sample (30 is the benchmark). Smaller sample size leads to bigger tails

Question 27

Power

Answer 27

The probability that study will produce a statistically significant result if the research hypothesis is true. Influenced by: Sample size - large ↑; small ↓ Pop 􏰀􏰀σ - small ↑; large ↓ (more variance) Significance level - high α (0.05/0.10) ↑; low α (0.01/0.001) ↓ # Tails: one-tailed ↑ ; two-tailed ↓

Question 28

Steps in the Scientific Method (6)

Answer 28

Draw conclusions and publish results

Question 29

Confidence Intervals

Answer 29

The range of scores that is likely to include the true population mean. The range of possible population means from which it is not highly unlikely you could have obtained your sample mean