Exam #1

Question 1

Measurement

Answer 1

is the process of collecting information from a test. The raw scores from a test.

Question 2

Evaluation

Answer 2

is the use of measurement in making decisions

Question 3

Public relations

Answer 3

proof that your program is successful and worthwhile

Question 4

Formative vs. Summative evaluation

Answer 4

Formative – Progressive and developmental

Summative – Terminal (e.g., letter grade, pass/fail)

Question 5

Data

Answer 5

Set of raw scores

Question 6

Raw score

Answer 6

An individual score
Indicate the time, distance, weight, etc.
One raw score cannot necessarily be directly compared to another raw score, especially if they are from different events or types of performance
A standard score can be calculated to compare different
sets of data to one another

Question 7

Nominal scale

Answer 7

Simplest, least precise measurement scale. Ex) jersey #, position

Question 8

Ordinal scale

Answer 8

More precise than nominal as it has order. Implies “more than” or “less than”. Ex) 1st, 2nd, etc. Does not allow calculation to be made because ordinal difference doesn’t imply equal differences.

Question 9

Interval scale

Answer 9

– More precise than nominal and ordinal

data. Ex) 20C is warmer than 10C however it is not twice as warm

Question 10

Ratio scale

Answer 10

The most precise and useful level of measurement. The ratio scale is essentially the interval scale with an absolute zero that indicates an absence of the measured attribute.

Question 11

Discrete variable

Answer 11

Specific or exact values (whole #’s)

Question 12

Continuous variables

Answer 12

Values on a continuum, can include decimal values. Ex) body composition values
Ex) 0.1sec vs 0.095 sec on an agility test

Question 13

________ is a mathematical representation of the distribution of data

Answer 13

The normal curve

Question 14

For normally distributed interval/ratio data, _______ are distributed around the____ in a symmetrical pattern

Answer 14

Raw scores, mean

Question 15

Scores cluster around the ____ or the ____

Answer 15

average, mean

Question 16

The highest peak represents the ____, ____, and _____

Answer 16

Mean, median and mode

Question 17

Normal distribution values

Answer 17

0.13%, 2.15%, 13.59%, 34.13%, 34.13%, 13.59%, 2.15%, 0.13%

Question 18

Common standard scores are:

Answer 18

Percentile rank, z-score, and t-score

Question 19

Percentile rank is an ____ ______ of data compared to some norm

Answer 19

Ordinal measurement

Question 20

Percentile rank

Answer 20

Provides quick comparison to the rest of the data

Good, but limited used; they are terminal statistic

(Equal percentile rank difference doesn’t mean equal performance difference; not effective for evaluating performance change or improvement)

Question 21

Formula for percentile rank using nearest rank method

Answer 21

N = ((100-P)/100) X N + (1/2)

Question 22

Calculation percentiles for data

Answer 22

P= 100((i-0.5)/n)

p= percentile 
i= rank 
n= number of raw scores

Question 23

Z-Scores

Answer 23

A standard score expressed in terms of standard deviation units on a normal curve.
The higher the score the further from the mean
A negative score means below the mean

Question 24

Z-Score formula

Answer 24

Z= (particular data value-standard deviation) / mean

Question 25

Mean

Answer 25

Average of a set of scores | take all data and divide by how many there is

Question 26

Median

Answer 26

Middle point of scores. 50% fall above or below this point. Order all points and find the one in the middle.

Question 27

Mode

Answer 27

Most frequently occurring. Order and then see which number repeats, if no answer then leave the square blank.

Question 28

Improvement

Answer 28

Evaluation of instruction & programs | Compare test results of a sample to national averages or norms

Question 29

Motivation as a testing tool

Answer 29

Testing process can be used as a motivational tool if done well and should be a positive process (e.g., KBB Christmas Break)

Question 30

Test-Retest

Answer 30

Same subjects, same conditions, tests conducted on two separate occasions should lead to the same result

Question 31

Alternate form

Answer 31

Two different versions of the same test should yield the same result (e.g., vertical jump test with either the vertec or jump mat)

Question 32

Correlations are used to determine?

Answer 32

Describe the strength of the relationship between the variables

Question 33

Are correlations always cause and effect?

Answer 33

Not necessarily

Question 34

Reliability, validity and objectivity scale for excellent, high, average and unacceptable.

Answer 34

``` Excellent = +.90 to +1.00 High = .80 to .89 Average = .60 to .79 Unacceptable = .59 ```

Question 35

Regression

Answer 35

Takes it one step further than correlation | Determines the formula for the relationship between the variables

Question 36

The stronger the correlation between two variables?

Answer 36

The better the regression equation will be at predicting

Question 37

T-Test for independent groups

Answer 37

2 different groups that are unrelated

Question 38

T-Test for dependent groups

Answer 38

Same group, tested twice, pre-test and post-test

Question 39

Accept Ho

Answer 39

no significant difference (means are the same) Since our calculated value is LOWER than the table value, we must ACCEPT the null hypothesis

Question 40

Reject Ho

Answer 40

– there is a significant difference (one group is better than the other or one training method is better than the other) Since our calculated value is HIGHER than the table value, we must REJECT the null hypothesis

Question 41

Inferential statistics

Answer 41

is testing the characteristics and abilities of a sample and generalizing about the entire population. (As an example -taking a sample of 25 out of 150 Kinesiology students, testing these 25, and then using their results to make predictions or assumptions of the entire group that they would have similar characteristics. The sample of 25 would have to be representative of the entire population.)

Question 42

Empirical research

Answer 42

based on observed and measured phenomena and derives knowledge from actual experience rather than from theory or belief. Data from Empirical research is called empirical data.

Question 43

3 ways the median is described

Answer 43

The second measure of central tendency The middle point of an ordered distribution of scores 50% of the scores fall above and below this point

Question 44

What does K mean in Anova test?

Answer 44

the # of groups in total

Question 45

F value

Answer 45

we use to determine whether we are rejecting or accepting the null

Question 46

Objectivity

Answer 46

The degree to which different administrators of a test assign or give scores that are the same.

Question 47

Reliability

Answer 47

The degree to which a test measures whatever it measures, consistently.

Question 48

Validity

Answer 48

The ability of a test to accurately measure the attribute it is designed to measure

Question 49

Content validity

Answer 49

contains elements that match what you are testing (speed test involves sprinting)

Question 50

Criterion-related validity

Answer 50

a test that correlates strongly to what you are measuring (aka concurrent - beep test vs metabolic cart)

Question 51

Construct validity

Answer 51

measuring more abstract or complex concepts that are tough to directly measure

Question 52

Descriptive Stats

Answer 52

testing all group members and then doing calculations on them

Question 53

Correlation +1.00

Answer 53

shows a perfect correlation which means a perfect positive relationship or correspondence (all dots are co-linear)

Question 54

Correlation -1.00

Answer 54

-1.00 shows a perfect negative correlation, which means there is an inverse relationship between the two sets of scores. (all dots are co-linear)

Question 55

Correlation 0.00

Answer 55

0.0 indicates no correlation or relationship (flat line with no slope)