Psyc 301 Exam 1

Data

information collected from the sample on the variable we are interested in

Nominal scales

measures that split people into categories - categories must be mutually exclusive Nominal = Nameable

Validity

measuring exactly what you intend to measure

inter-rater reliability

A measure of how similarly two different test scorers would score a test.

What are correlations?

A measure of the relationship between two variables

Why is it important?

A set can have the same sample size and the same mean by have a different amount of variability

Bar graphs vs histograms

Bar graphs- compare the frequency of categorical responses Histograms- compare the distribution of continuous variables

Misleading graphs

Beware the scale of the axes

Goals of science and how statistics help achieve those

Description- How do people behave? Prediction- Identifying the factors that influence behavior Explanation-Identifying the underlying causes of a behavior Research methods and statistics are the ways we learn and discover

Correlation vs. Causation

Correlation does not equal causation 1. Reverse Causation: may flow from Y to X 2. Reciprocal Causation: two variables cause each other 3. Common Causal Variables: third variable influence the other 2 variables

Variability

The extent to which the scores in a data set tend to vary from each other and from the mean. -Range -Standard Deviation -Variance

Reporting correlations

There was a strong positive correlation between age and happiness (r=.61), suggesting that as age increases, so does happiness.

When to use which measure of central tendency

Use MODE when the data are categorical Use MEAN when the data are continuous and you don't have any extreme scores Use MEDIAN when the data are continuous and you think mean is misleading -When in doubt report both

Continuous Data

data measured on a continuum, all numbers between two end points are possible scores - Ex: height, weight, age

Categorical Data

data that sorts people into categories, only so many options for the variable - Ex: gender, major, experimental condition

Bimodal

a data set with two modes

Scatterplots

a graphed cluster of dots which represents the values of two variables

Statistics

a set of tools and techniques used for describing, organizing, and interpreting information or data

Central Tendency

a single number number that represents a group of scores -Mean -Median -Mode

Identifying Extreme Values

any data point more than 2 SD away from the mean is a potential outlier

Limitations of the correlation coefficient

can only identify LINEAR relationships

parallel forms reliability

consistency between alternate versions of the same test

reliability

consistency of a measure

error score

discrepancy between observed score and true score

Kurtosis

how flat or peaked a normal distribution is - Platykurtic: low kurtosis, flat more variability - Leptokurtic: high kurtosis, peaked less variability

coefficient of determination

how much variation 2 variables share -r^2

Measurement scales

nominal, ordinal, interval, ratio

Ordinal scales

number is in a ranking - unclear how much distance separates the data Ordinal = ordering

Pearson correlation

only used for 2 continuous variables - most common type

Interval scales

ordered events with equal spacing - zero not necessarily meaningful Most common type

Direction of a Correlation

positive or negative -positive relationship: direct relationship, same direction -negative relationship: indirect relationship, opposite direction

Computing a correlation coefficient

r=ΣXY-n̅x̅y̅/√(ΣX²-n ̅x̅²)√(ΣY²-n ̅y̅²) -Do numerator and denominator separately in order to prevent mistakes. -for r, it doesn't matter which variable is X or Y because each variable goes through the exact same process in the formula.

Skewness

refers to the lack of symmetry -positive skewness (right foot) -negative skewness (left foot)

Percentile Points

refers to the percentage of cases equal to and below a certain point in a group of scores -The median is the 50% percentile

Cronbach's alpha

reflects the degree of internal consistency

discriminant validity

scores on the measure are not related to other measures that are theoretically different

convergent validity

scores on the measure are related to other measures of the same construct

Ratio scales

similar to interval scale but 0 has specific meaning - zerO = ratiO uncommon

Correlation matrix

simple way to report multiple correlations

Variable

something that can change or have different values for different individuals

Variance

standard deviation squared -s^2

Measures

the act or process of assigning numbers to phenomena according to a rule - Behavioral measures - Self-report measures - Physiological measures

coefficient of alienation

the amount of unexplained variance

Standard Deviation

the average amount of variability in a set of scores - most common measure of variability - low SD data points are close to sample mean - high SD data points are far away from sample mean s = √[(Σ(x - xbar)^2)/n-1]

Mean

the average value of a group of numbers - most common X-bar = (Σx)/n

Strength of a Correlation

the closer the absolute value of the correlation is to 1 the stronger the relationship -Very strong .8 to 1 -Strong .6 to .8 -Moderate .4 to .6 -Weak .2 to .4 - No to weak 0 to .2

internal consistency (inter-item reliability)

the degree to which a test yields similar scores across its different items

Range

the difference between the highest and lowest scores in a distribution - most general measure of variability - ignores the middlemost values - emphasis on extreme scores r = h - l

criterion validity

the extent to which a measure is related to an outcome

predictive validity

the extent to which a score on a scale predicts scores

content validity

the extent to which a test samples the behavior that is of interest

concurrent validity

the extent to which two measures of the same trait or ability agree

construct validity

the extent to which variables measure what they are supposed to measure or don't measure what they shouldn't

Sample

the group you actually collect data from

Population

the group you are actually interested in drawing some conclusions about

Histograms

the height of each bar is the number of times each value occurs in our data set -allows us to see the distribution of our data

Median

the midpoint in a set of scores -point at which half of the scores are bigger and half of the scores are smaller List values in order and find middlemost score

observed scores

the score you actually got

Mode

the value that occurs most frequently in data - typically used with categorical data

true scores

true reflection of what you really know

Inferential Statistics

used to make inferences about a larger group from a smaller group - The next step after description

Descriptive Statistics

used to organize and describe data - Ex: counts, means, percentenges

test-retest reliability

using the same test on two different occasions to measure consistency