Chapter 13: Descriptive Data Analysis

Front 1

why is analysis of data so important?

Back 1

only by applying some analysis to the data can it be condensed to a point where it can be meaningful to all interested people in terms of overall performance or characteristics of the group. It is also important to make comparisons between groups.

Front 2

what are the three distinct steps in data analysis?

Back 2

1. select the technique appropriate for the data and the research question.

2. apply the technique or calculate using the technique.

3. interpret the result of the technique.

Front 3

why is step 2 best accomplished with the aid of a computer?

Back 3

because it is difficult when there is a large amount of data.

Front 4

sample information is used for what?

Back 4

sample information (statistics) are used to estimate population information (parameters).

Front 5

define statistics

Back 5

values or quantities calculated using information obtained from a sample; used to estimate population information.

Front 6

define parameters

Back 6

values or quantities calculated using information obtained from a population.

Front 7

define descriptive statistic

Back 7

statistics used to describe characteristics of a group.

Front 8

define continuous scores

Back 8

scores that have potentially infinite number of values.

Front 9

define discrete scores

Back 9

scores that are limited to a specific number of values.

Front 10

scores can classified as what?

Back 10

1. continuous

2. descrete

Front 11

most scores are in kinesiology are what kind of scores?

Back 11

continuous

Front 12

discrete score are normally not what?

Back 12

expressed in fractions

Front 13

scores can also be classified as what?

Back 13

1. ratio

2. interval

3. ordinal

4. nominal

Front 14

define ratio scores

Back 14

scores that have a common unit of measurement between each score and a true zero point. statements about equality of ratios can be made.

Front 15

define interval scores

Back 15

scores that have a common unit of measurement between each score, but do not have true zero point.

Front 16

define ordinal scores

Back 16

scores that do not have a common unit of measurement between each score, but are ordered from high to low.

Front 17

define nominal scores

Back 17

scores that cannot be hierarchically ordered.

Front 18

how should scores be recorded in order to analyze it?

Back 18

they must be recorded in a single unit of measurement (usually the smaller one).

Front 19

when should computer analysis be used?

Back 19

when the set of scores is large (50+). used to ensures accuracy and speed.

Front 20

what are some common computer programs packages for statistical computations?

Back 20

1. SAS

2. SPSS

3. Statview

Front 21

How can you organize and graph scores?

Back 21

1. Simple frequency distributions

2. grouping scores for graphing

Front 22

define simple frequency distribution

Back 22

an ordered listing of the scores and their frequencies.

Front 23

what does the interval size when grouping scores tell us?

Back 23

tells us how many scores to group together

Front 24

Interval size

Back 24

interval size = lgest score - smallest score/15

Front 25

define frequency polygon

Back 25

a line graph of scores and their frequencies

Front 26

define normal curve

Back 26

a symmetrical curve centered around a point with a defined base to height ratio, indicating a balanced (or normal) distribution; also called a bell-shaped curve.

Front 27

define skewed curve

Back 27

an asymmetrical curve indicating an unbalanced distribution. Can be positively skewed or negatively skewed.

Front 28

define leptokurtic curve

Back 28

a curve that is more sharply peaked than a normal curve, indicating an extremely homogenous group.

Front 29

Platykurtic curve

Back 29

a curve that is less sharply peaked than a normal curve, indicating an extremely heterogenous group.

Front 30

define histogram

Back 30

a bar graph of scores and their frequencies. In a there are about 10 scores (interval midpoints).

Front 31

what happens once a large set of scores has been collected?

Back 31

certain descriptive values can be calculated; descriptive value will summarize or condense the set of scores and give it meaning.

Front 32

what are descriptive values?

Back 32

they are used by researchers primarily to describe the performance of a group or compare it's performance with that of another group.

Front 33

List the types of descriptive values.

Back 33

1. Measures of central tendency

2. Measures of variability

Front 34

what are the measures of central tendency?

Back 34

1. Mode

2. Median

3. Mean

Front 35

define central tendency

Back 35

a descriptive value that indicates those points at which scores tend to be concentrated.

Front 36

define mode

Back 36

a measure of central tendency used with nominal data that indicates the score most frequently received.

Front 37

define median

Back 37

a measure of central tendency used with ordinal data that indicates the middle score.

Front 38

define mean

Back 38

a measure of central tendency, used with interval or ratio data, that is the sum of the scores divided by the number of scores.

Front 39

What are measures of variability?

Back 39

describes the data in terms of their spread or heterogenity.

Front 40

what are the types of measure of variability?

Back 40

1. Range

2. Standard deviation

Front 41

define range

Back 41

a measure of variability that is the difference between the highest and lowest score.

Front 42

define standard deviation

Back 42

a measure of variability that indicates the amount that all the scores differ from the mean. The more the scores differ from the mean, the larger the standard deviation.

Front 43

how can you tell if a score is normally distributed?

Back 43

few, if any scores will be more than three standard deviations away from the mean.

Front 44

define outlier

Back 44

an extremely high or low score that does not seem typical for the group tested. Often when score is more than thee SDs away from the mean it is an outlier.

Front 45

define variance

Back 45

a measure of variability that is the square of the standard deviation. it is not a descriptive statistic, but is a useful statistic in certain other statistical procedures and interpretations.

Front 46

list some way for measuring group position (position or rank of participants within a group based on scores)

Back 46

1. Percentile ranks and percentiles

2. standard scores

Front 47

when are percentile ranks and percentiles often calculated?

Back 47

when developing norms.

Front 48

what is a disadvantage of percentile ranks?

Back 48

they are ordinal scores and there is not common unit of measure between consecutive percentile rank values because they are position measure.

small changes in score values near the mean result in large changes in the percentile ranks.

Front 49

are percentile ranks/percentile packages with SPSS?

Back 49

no but there are computer programs available for percentiles and percentile ranks.

Front 50

when do researchers use standard scores?

Back 50

when they have data on participants from several different tests, and they want all the data in the same unit of measurement. Each test score is converted to a z-score.

Front 51

define standard scores

Back 51

scores that are used to change the scores from different tests to a common unit of measure. For each test, test scores are converted to z scores by using a formula.

Front 52

what is the formula for standard score?

Back 52

z= test score - mean for test/SD for test

Front 53

define z score

Back 53

a standard score with mean 0 and standard deviation 1.0. A z score indicates how many standard deviations a test score is from the mean. Typically a z score is between -3.0 and 3.0 and are fractional.

Front 54

True of false: most pacakages of statistical programs have a transformation feature that could be used to calculate new score (like z scores) using scores already entered into the computer.

Back 54

true

Front 55

to determine if there is a relationship between two variable, what must be true?

Back 55

every participant must be measured on each variable.

Front 56

how many techniques are there to determine a relationship? What are they?

Back 56

two

1. graphing technique

2. mathematical technique

Front 57

what are the differences between the graphing technique and mathematical technique?

Back 57

graphing technique is not as precise, but prior to computers it was easier and quicker.

Front 58

how does a computer graph a relationship?

Back 58

the computer develops a coordinate system with values of one variable listed along the horizontal axis and values of another variable listed along the vertical axis. It plots a point for each participant above their score on the horizontal axis and oppossite their score on the vertical axis. The graph that results is called a scattergram.

Front 59

define scattergram

Back 59

a graph that shows the relationship between two variables.

Front 60

define line of best fit

Back 60

a straight line that represents the trend or relationship in the data; also known as the regression line.

Front 61

what does the line of best fit represent?

Back 61

represents the trend in the data.

Front 62

positive relationship

Back 62

when large scores on one measure are associated with large scores on the other measure.

Front 63

negative relationship.

Back 63

when large scores on one measure are associated with small scores on the other measure.

Front 64

true or false: the further the plotted points to the trend line, the higher or larger the relationship.

Back 64

false: the closer the higher or larger the relationship.

Front 65

what does it mean when the plotted points resemble a circle or it is impossible to draw a trend line?

Back 65

there is no linear relationship.

Front 66

what is the correlation technique?

Back 66

mathematical technique for determining the relationship between two sets of scores.

Front 67

who developed the formula for correlation technique?

Back 67

Karl Pearson

Front 68

define correlation

Back 68

a mathematical technique for determining the relationship between two sets of scores.

Front 69

correlation coefficient (r)

Back 69

a value indicating the degree of relationship between two sets of scores.

Front 70

what are the two characteristics of correlation coefficient?

Back 70

1. direction (indicates whether correlation coefficient is positive or negative.

2. Strength indicates how close r is to 1.

Front 71

what shows a perfect positive relationship?

Back 71

r = 1

Front 72

what shoes a perfect negative relationship

Back 72

r = -1

Front 73

define rho

Back 73

a value indicating the degree of a relationship between two sets of ranks; also known as Spearman's rho and rank order correlation coefficient.

Front 74

what does a high correlation between two measures indicate?

Back 74

it does not usually indicate a cause-and-effect relationship.

Front 75

what is the true indicator of the degree of relationship?

Back 75

coefficient of determinination

Front 76

define coefficient of determiniation

Back 76

a value that indicates the amount of variability in one measure that is explained by the other measure. it is the the square of the correlation coefficient (r2)

Front 77

define linear relationship

Back 77

a relationship between two variables best represented graphically by a straight line.

Front 78

define curvilinear relationship

Back 78

a relationship between two variables best represented graphically by a curved line.

Front 79

true or false: a linear relationship is best shown graphically by a straight line, as the trend line, however a relationship between two sets of scores may be best represented by a curved line, indicating a curvilinear relationship.

Back 79

true

Front 80

define relative risk (RR)

Back 80

likelihood that group with a risk factor will have a health characteristic in comparison to a group without a risk factor;

Front 81

define odds ratio (OR)

Back 81

an estimate of relative risk

Front 82

define attributable risk (AR)

Back 82

percentage of a cases in the total group that occur in the group with a risk factor.