Final Exam Statistics.

The numerator of an inferential test statistic

-The actual difference between means

The denominator of an inferential test statistic

-the average distance between means -the distance between M and u

Standard error is determined by

1. Variability of raw scores (Standard Deviation) 2. Size of the sample (n)

Histogram or Polygraph

Interval and Ratio data

Why we use computational formua

It works with decimals

Inferential Statistics

Hypothesis testing, predicting from a sample to make a prediction on the population

Factor

Independent Variable

Distribution of F ratio is

positively skewed

Critical Region

Region of values that corresponds to the rejection of the null hypothesis

When is sampling error a problem?

When a sample is selected that doesn't represent the entire population

When is the median the most appropriate central tendency?

When the data is skewed

What does the sign of the z-score tell us?

Whether the value is above or below the mean

What values are possible for a F ratio?

Will always be positive

What does the absolute value of the z-score tell us?

How many standard deviations you are away from the mean

As sample size increases, standar error _______

Decreases

Degrees of Freedom

Describes how well the T statistic represents a Z score

Descriptive Statistics

Desribes, summarizes, organizes set of data, bar graph, mean, etc.

Nominal

Label and Categorize. Each group has no significance over the other EX. SSN

Variablility Definition

The spread of a distribution, whether clustered together or spread apart

Central Limit Theory

Used to specify the shape, central tendency and variability of the distribution of sample means

Single Sample Z Test

Used when comparing means, the known values are the population mean and SD

Single Sample T Test

Used when comparing means, the only known value is the population mean

Ratio

Absolute zero point. EX. Age, Height, Weight

Value of Correlations

Between -1.00 and 1.00

Quasi-Experiement vs Correlational

Both: Nothing is being manipulated Correlational: Same group of people, comparing two variables. Quasi: Comparing two separate groups

Related T Test

Comparing one group of people, 2 variables each

Independent T Test

Comparing two groups of people, one variable each

The T Statistic

Uses sample data to test hypotheses about an unknown population mean

When is it not necessary to use statistics to draw conclusions about a population?

Descriptive statistics uses the data to provide descriptions of the population, either through numerical calculations or graphs or tables. Inferential statistics makes inferences and predictions about a population based on a sample of data taken from the population in question.

Standard Deviation

Distance from a X raw score to a mean

Effect Size _____ change with sample size

Does NOT

Chi Square test

Examining relationships, Nominal data

Regression

Examining relationships, making a predicition based on knowledge from another variable

Hypothesis Testing

Goal is to rule out chance (sampling error) as a explanation for results

Sampling Distribution

How we estimate sampling error. It's a distribution, not a score

A larger smaple creates ______ discrpency (Sample Error)

Less

Which measure of central tendency can only be used for desriptive statistics?

Mode

Interval

Numerical Ordered categories, EX. Temperature

Bar Graph

Ordinal and Nominal Data

True experimental designs vs quasi experimental designs

Quasi- Experimental designs do not manipulate anything, and no random assignment

Three sources of Variance Discussed

Sum of Squares Variance Standard Deviation

What does the standar deviation measure?

Variability

Ordinal

categories with some order. Ranking. EX. 1st, 2nd, 3rd

Sampling Error

discrepancy between sample statistic and population parameter

Standard Error

distance from sample mean to population

Finding Probability

draw normal distribution, label mean and SD, identify and shade the area of interest. (Body, Tail, or Mean to Z)

Type II Error

false negative

Type I error

false positive

Experiment wise error

if we do multiple t tests, type I error accumulates

Purpose of Chi Square Test

intended to test how likely it is that an observed distribution is due to chance.

Variability of the sampling distribution

measured by standard error of mean

Probability

method for measuring the likelihood of obtaining a specific sample from a specific population