# It's All About Variables!

A condensed text for Foundations of Statistics for the global learner.

- Created by: Debra Wilson
- Created on: 23-10-12 01:32

First 28 words of the document:

It's All About Variables! 1

It's All About Variables!

Re-written by a global thinker

who was driven nuts by the text book.

## Other pages in this set

### Page 2

Here's a taster:

It's All About Variables! 2

Table of Contents

Population versus Sample...................................................................................................................................3

Selecting Samples...............................................................................................................................................3

Simple Random Sample (SRS)......................................................................................................................3

Stratified Random Sampling..........................................................................................................................3

Cluster Sampling............................................................................................................................................3

Systematic Sampling......................................................................................................................................3

Convenience Sampling..................................................................................................................................3

Self-Selected Samples....................................................................................................................................3

Bias..........................................................................................................................................................................4

VARIABLES Things that vary.............................................................................................................................5

Experiment Design Observational or Experimental........................................................................................5

NUISANCE VARIABLES.................................................................................................................................5

CONTROLLING SUBJECT NUISANCE VARIABLES.............................................................................5

CONTROLLING SITUATION NUISANCE VARIABLES.........................................................................6

CONFOUNDING FACTORS........................................................................................................................6

FAULTS.........................................................................................................................................................6

What's Normal?.......................................................................................................................................................7

Standardised Values (z-scores)...........................................................................................................................7

Significance Theory............................................................................................................................................8

P value - Significance ....................................................................................................................................8

From Population to Samples The Proportion..............................................................................................8

Tests for Categorical Variables..............................................................................................................................10

REPORT...........................................................................................................................................................…read more

### Page 3

Here's a taster:

It's All About Variables! 3

Population versus Sample

Populations are all the things we are interested in; they can be all people, cars, cats or toothpicks. We can be

interested in more than one population at the same time.

Samples are a random selection of a proportion of the population used to test the hypothesis or research

question against.…read more

### Page 4

Here's a taster:

It's All About Variables! 4

Bias

Who organised the study?

· Was the study is organised by a company or people with a vested interest in getting a certain result.

e.g. CSR organising a study on whether too much sugar is bad for your health.

Who conducted the study?

· Was the study conducted by a company or people who have a vested interest in getting a certain

result. e.g. Out of work sugar producers trying to get a factory re-opened conduct the study.…read more

### Page 5

Here's a taster:

It's All About Variables! 5

VARIABLES Things that vary.

Categorical can contain any value from a range of values. e.g. Sex = F, M (or 0 for F and 1 for M), range of

10,20,30,40 = each time it can contain any one of these numbers.

Metric can contain only one numeric value only e.g. Score = 35.6, How much do you weigh in Kg's = 84.

Variables can be either Independent (IV) or Dependent (DV).

CHANGE EFFECT

relationship

IV DV

e.g.…read more

### Page 6

Here's a taster:

It's All About Variables! 6

Matched Pairs Design

use pairs of participants carefully matched on all major subject nuisance variables

for each pair, one is randomly assigned to each of the treatment conditions

use this avoids practice effects, but

you need to identify all major nuisance variables and find several carefully matched pairs, e.g.…read more

### Page 7

Here's a taster:

It's All About Variables! 7

What's Normal?

Example: 70 out of 1000 said yes.

Proportion = Frequency/ Total Number

70 / 1000 = 0.07

Percentage = proportion x 100

0.07 x 100 = 7%

The normal distribution is described as a "symmetric, bell shaped curve". For any given mean and standard

deviation you can draw a normal curve, and regardless of what the mean and standard deviation happen to

be, 95% of values fall within two standard deviations of the mean.…read more

### Page 8

Here's a taster:

It's All About Variables! 8

Significance Theory

P value - Significance

Tells us what the chance is of getting a proportion as high/low in the sample if the proportion was still the

known proportion.

0.1 = 1 chance in 10, 0.01 = 1 chance in 100, 0.001 = 1 chance in 1000

From Population to Samples The Proportion.

In the above histogram and percentiles chart the sample proportions

cluster around .30 and 50% of sample proportions fall between .25

and .33.…read more

### Page 9

Here's a taster:

It's All About Variables! 9

EXERCISE 3.2 from Foundations of Statistics, 2nd Edition

In 2006, 62% of all AFL players lived in Victoria. Suppose we took a random sample of 25 AFL players from this

population.

We would expect the sample

proportion of AFL players who

lived in Victoria to be somewhere

around 0.62, but not necessarily

exactly equal to 0.62.

Each block represents the

proportion of players who live in

Victoria in one random sample of

25 AFL players.…read more

### Page 10

Here's a taster:

It's All About Variables! 10

Tests for Categorical Variables

Describing the distribution of a single categorical variable.

SPSS Analyze/Descriptive Statistics/Frequencies

Use Pie chart for 2 or 3 categories

Use Bar chart for 3 or more categories

What you need to know.

Total of participants in sample = 416. Always use the valid percent column for percentages.…read more

## Comments

No comments have yet been made