Research Methods Final

Front 1

Raw Data

Back 1

Numerical data collected from each participant

Front 2

Dataset

Back 2

Collection of the raw data for the same variables for a set of participants

Front 3

Statistics

Back 3

Any numerical indicator of a set of data
2 types
Descriptive and Inferential

Front 4

Descriptive Statistics

Back 4

Convey essential information about the data as a whole; simple descriptions about characteristics of a set of quantitative data
3 Types
Frequency
Measures of Central Tendency
Measures of Dispersion

Front 5

Inferential Statistics

Back 5

Help us draw conclusions about the population of interest via the sample we took. also helps us understand relationships between variables

Front 6

Normal Distribution

Back 6

theoretical distribution of scored where one side is a mirror image of the other side

Front 7

Positive Skew

Back 7

Many people below average, but few high scoring outliers make average higher than most people

Students hate this
Mean > Median

Front 8

Negative Skew

Back 8

many people score highly but a few low scoring outliers make the average lower than most

Students love this
Mean < Median

Front 9

Leptokurtic Distribution

Back 9

scores cluster tightly around the mean

Front 10

Platykurtic Distribution

Back 10

scored are less tightly clustered around the mean

Front 11

Mode

Back 11

Score that appears most often
can see bimodal or multimodal distributions, making it impossible to use mode to represent average

Front 12

Median

Back 12

splits dataset exactly in half, not swayed by outliers
better measure of central tendency in skewed distributions

Front 13

Population Distribution

Back 13

Frequency with which cases that make up a population are arranged (mean, median, mode, variance, Sd)

Front 14

Sample Distribution

Back 14

Frequency with which cases that make up a sample are arranged (statistics, Mean, Median, mode variance, Sd)

Front 15

Sampling Distribution

Back 15

frequency with which values of statistics are observed or expected to be observed when numerous random samples are drawn from a given population

Front 16

Central Limit Theorem

Back 16

if samples are drawn from a population at random their means tend to be distributed normally the bigger the sample size, the more likely this is
Grand Mean=Sample Mean=Pop. Mean

Front 17

Standard Error

Back 17

Index of sampling error (inaccuracy)- how far off our sample mean is from the true population mean
Standard Error:
Standard deviation/(squareroot of sample size)

Front 18

Confidence Interval

Back 18

Range of scores of random sample means associated with a confidence interval
How well our sample statistic is estimating the population parameter

Front 19

Systematic Error and Bias

Back 19

Sampling techniques, survey design

Front 20

Significance Levels

Back 20

The level of error the researcher is willing to accept for a given statistical test, established prior to conducting analysis

Front 21

Directional Hypothesis

Back 21

Region of rejection on one side of the distribution based on critical value. Null hypothesis includes no difference and a relationship in the opposite direction

Front 22

Non-Directional Hypothesis

Back 22

Region of rejection divided onto both sides of the distribution based on critical value, stricter b/c area of rejection is smaller
null hypothesis includes no difference or relationship

Front 23

Type I Error

Back 23

When researcher rejects a null hypothesis when it is probably true of should have been accepted
False positive
To Control:
lower significance level (p=.05 to p=.01)

Front 24

Type II Error

Back 24

When researcher accepts a null hypothesis when it's probably false and should have been rejected. can be reduced by increasing sample size.
False negative
To Control:
Raise significance level, increase sample size, make a directional hypothesis

Front 25

Observed Frequency

Back 25

Number of people in a sample that actually appear in a given categoryr

Front 26

Expected Frequency

Back 26

Number of people in a sample that we'd expect to appear in a given category based on a population view

Front 27

Limitations of Chi-Squared

Back 27

Categorical data only, Bivariate test will tell us that differences exist, but doesn't identify which cells are significantly different,
no way to determine causal relationships

Front 28

T-test

Back 28

when comparing 2 groups
difference between group A and group B
one variable
1 or 2 tailed

Front 29

Correlation

Back 29

Relationship, connection, correlation,
2 variables, one group of people
1 or 2 tailed

Front 30

Chi-Squared

Back 30

table,
counts of people, categorical
bivariate chi-squared needs a table

Front 31

Z Score

Back 31

Standardized distribution,
Deviation/(squared deviation)