Chapter 9 - Statistics Flashcards

1
Q

What two things do statistics inform about?

A
  1. Reliability

2. Meaningfulness

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

What are the most fundamental components of most statistical techniques?

A

Measures of central tendency and variability.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

What does the nominal level of measurement provide info about?

A

About difference, but not much more. It is used to name, identify, or classify into categories.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

What does the ordinal level of measurement provide info about?

A

Shows direction of difference, but we do not know amount of difference. Numbers indicate rank or order, allow for greater or less than.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

What does the interval level of measurement provide info about?

A

The intervals or distances between numbers, but it is not known how far any of the numbers are from zero. There is an equality of units, but no true zero.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

What does the ratio level of measurement provide info about?

A

Each number can be thought of as a distance measured from zero.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

What does the absolute zero point in the ratio level of measurement represent?

A

It represents the absence of the variable being measured.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

What is non-parametric data?

A

Data that does not meet the assumption of normality.

i.e. Nominal, ordinal levels.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

What is parametric data?

A

Data that does meet the assumption of normality.

i.e. Interval, ration levels.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

What does variability estimate and what is its formula?

A

It is the best estimate of the spread of scores.

s^2 = The sum of ((X - Xav)^2) / (N - 1)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

What are the levels of a normal distribution?

A

+1 sd = 68%
+2 sd = 95%
+3 sd = 99%

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

What does positive skewness look like?

A

Peak near the y-axis.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

What does negative skewness look like?

A

Peak far away from the y-axis.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

What does kurtosis represent?

A

Peakedness of distribution.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

What does the leptokurtic distribution look like?

A

Strong peak.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

What does the platykurtic distribution look like?

A

No peak, close to even distribution throughout.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

What does correlation measure?

A

A measure of the strength of the relationship between two variables, X and Y.

18
Q

What does the pearson product-moment correlation assume?

A

That the X and Y variables are normally distributed and are interval or ratio scale scores.

19
Q

What is the spearman rank-order correlation used for?

A

It is used for ordinal scale data, or interval or ration scale data that deviate substantially from normality.

20
Q

What is the pearson r relationship independent of?

A
  1. The # of scores.
  2. The size of scores.
  3. The dispersion of scores.
21
Q

What is pearson r derived from?

A

Covariance.

22
Q

What is covariance?

A

How much the deviations in the mean of your X variable related to the deviations in the mean of your Y variable.

23
Q

Null Hypothesis:

A

The observed correlation is due to error.

24
Q

When is r significant?

A

When it meets or exceeds the tabled critical value of r.

25
Q

What is statistical significance based on?

A

Largely based on sample size.

26
Q

When are one-tail significance levels used?

A

When the direction of the correlation has been predicted in advance.

27
Q

What does causation depend on?

A

Depends on methodology, not analysis.

28
Q

One Sample t-test:

A

Comparing one sample to the population. You need to know the population mean.

29
Q

Chi-square goodness of fit test:

A

When your data is not normally distributed but you still want to compare it to the population.

30
Q

Independent t-test:

A

Compare two independent samples (between factor).

Has a normal distribution, also called a two sample t-test.

31
Q

Dependent t-test:

A

Compares scores from the same or matched sample (within factor). One observation does depend on the other. Has a normal distribution.

32
Q

Mann-Whitney U:

A

When the data is skewed, not normally distributed.

Non-parametric independent sample test.

33
Q

Wilcoxon t-test:

A

Want to look at pre and post test but the data is not normally distributed.
Non-parametric dependent sample test.

34
Q

ANOVA:

A

Analysis of variance.

35
Q

One-way ANOVA:

A

One independent variable, normally distributed.

36
Q

Repeated measures one-way ANOVA:

A

Repeatedly measuring one group over time. Normally distributed.

37
Q

2 x 2 Split-Plot Factorial ANOVA:

A

Two independent variables. Between factor, time factor.

38
Q

Randomized factorial ANOVA

A

All groups are independent of each other. There are two IV. No within factors.

39
Q

MANOVA:

A

Has more than one DV.

40
Q

ANCOVA or MANCOVA:

A

Allows you to control for a third variable.

41
Q

What is regression used for?

A

Prediction and/or explanation