HOW TO USE (AND MISUSE) STATISTICS
by Gregory A. Kimble. Prentice-Hall Inc., 1978

PREFACE (pxi)

1) The nature of statistics (p1-19)

[1] Descriptive statistics (p2-7)

(1) Sizes of cities and how they grow (p3-4)

(2) Unemployment (p4-6)

(3) Rates of mental illness (p6)

(4) Crime rates (p7)

[2] Operationism and some related ideas (p7-9)

(1) Operationism and common sense (p8-9)

(2) Settling arguments (p8)

It is very easy for arguments to arise regarding the sizes of things. When determining "rates" of change, for example, it is all a question of the "bases" used for the comparisons. It depends upon the "definition" agreed upon of what constitutes the basis. If the same definition is NOT accepted for the thing being argued about, at least it should be understood that the argument is about definitions rather than about the facts of the situation!

(3) Silly statements --- For example, the assertion "People never use 100% of their "brainpower" but only about 10% of it" is operational nonsense! Why? (p9)

Because ---

1. How do you define "brainpower?"

2. What kind of measurement is involved that will permit quantitative statements?

3. If 100% of your brainpower is never used, how could you possibly know that you only use about 10% or some small amount of it? To know the value of a fraction of something demands that you know the value for the whole!

The new personalistic, dynamic, humanistic discussions of mental life are full of similar operationally silly statements. For example, the phrase "You must get in touch with your feelings" is often repeated at sales promotions or during counseling. It is a curious phrase because it flouts scientific examination. The phrase implies that one has feelings one does not feel!

4. If people in fact do function below their best brainpower levels --- which may be true, why blame it on brainpower rather than on poor motivation, sloppy work habits, or somthing else?

[3] Inferential statistics (p10-16)

(1) The riddle of the neglected lover (p10-11)

(2) Was it the fickle finger of fate? (p11-12)

(3) Making wise decisions in the face of uncertainty (p12-13)

(4) Types of error (p13-14)

(5) To put it briefly (p14-15)

Statistical ideas are not very difficult. Statistics is not so much about mathematics, formulas, and calculations as it is about ways of reasoning! Such statistical reasoning applied to making inferences from data (numbers of things or people) can be summarized in a series of five steps: (p15)

1. State the hypothesis in "null form."

2. Obtain data of the kind identified in the null hypothesis

3. Determine the chance probability of occurrence (4 or 5 times in 100) of the data obtained if the null hyhpothesis is true.

4. If the chance probability of the obtained result is swmall, reject the null hypothesis with a level of confidence that is the probability of obtaining the result by chance.

5. Recognize that either rejecting or accepting the null hypothesis involves a gamble.

(6) Research hypotheses (p16)

[4] Parameter estimation (p16-18)

[5] Summary --- Glossary (p18-19)

2) Pictures of data (p21-41)

3) Frequency distributions (p43-60)

4) The tasks of science (p61-83)

5) The laws of chance (p85-107)

6) The normal curve (p109-132)

7) Sampling the universe (p133-157)

8) Correlation (p159-178)

9) Uses and misuses of correlation (p179-200)

[1] Percentage of relationship" (p180)

[2] Correlation versus base-rate activity (p180-184)

[3] The calculation of "Heritability" (p184-186)

[4] Reliability and validity of measurement (p186-189)

(1) Reliability (p186-188)

(2) Validity (p188-189)

[p5] Validity and human justice (p189-192)

10) Anova (p201-220)

COMPUTATIONAL APPENDIX (p221-253)

END NOTES (p255-256)

INDEX (p287-290)

Go to: Brainpower Issue at Work
Go to: Change Issue at Work
Go to: Feelings Issue at Work
Go to: Leadership Issue at Work
Go to: Performance Issue at Work
Go to: Success Issue at Work
Go to: Leadership Control Essay
Go to: Home Page Index
Go to: Interactive Index