421 - Health Data Analysis

Front 1

Are different types of variables analyzed differently?

Back 1

Yes

Front 2

What are the two ways that data is organized and displayed?

Back 2

Numerically and Visually

Front 3

How is data displayed numerically?

Back 3

Through the use of tables.

Front 4

How is data displayed visually?

Back 4

Through the use of graphs and charts.

Front 5

How many variables does a frequency table display?

Back 5

One

Front 6

This contains the number and percentage of people who gave each response?

Back 6

Frequency Table

Front 7

Does a frequeny table contain the number of people for whom no response are available?

Back 7

Yes.

Front 8

You can find the cumulative percentage of a variable here.

Back 8

Frequency Table

Front 9

This is one of the most convenient and useful ways of summarizing data numerically.

Back 9

Frequency Table.

Front 10

Do quantitative variables (ratio or interval; age, degrees) work well in a frequency table?

Back 10

No

Front 11

What types of variables work best in a frequency table?

Back 11

Nominal and Ordinal (Categorical Variables)

Front 12

What way(s) are best for displaying data visually when you are analyzing a categorical variable?

Back 12

Pie charts and bar charts.

Front 13

What way(s) are best for displaying data visually when you are analyzing a quantitative variable?

Back 13

Histograms

Front 14

This is a graphical display of counts for a range of data values for one variable.

Back 14

Histogram

Front 15

This axis displays the range of sample values for a variable of interest.

Back 15

X-axis

Front 16

This axis displays the count (or frequency) of observations within a range of values (height of bar determines frequency of values in that range)

Back 16

Y-axis

Front 17

In a bar chart, each bar stands for how many response choices?

Back 17

One

Front 18

In a histogram, each bar stands for how many response choices?

Back 18

It varies - it represents a range

Front 19

What kind of information can you get from a histogram that you cannot get from a pie chart or a bar chart?

Back 19

1) Cluster around central value?
2) Outlying data points that are far removed from the rest?
3) Is distribution symmetrial or not (skewness)?

Front 20

If a skew exists to the right on a histogram, what is that also referred to as?

Back 20

A positive skew.

Front 21

If a skew exists to the left on a histogram, what is that also referred to as?

Back 21

A negative skew.

Front 22

For what type of variable would you use a pie chart?

Back 22

Categorical Variable (nominal or ordinal)

Front 23

For what type of variable would you use a bar chart?

Back 23

Categorical Variable (nominal or ordinal)

Front 24

For what type of variable would you use a historgram?

Back 24

Quantitative (ratio or interval)

Front 25

A frequency table can show us what two pieces of information that assists us in understanding quantitative variables?

Back 25

1) Measures of central tendency
2) Measures of variability

Front 26

What are the 3 measures of central tendency?

Back 26

1) Mean
2) Median
3) Mode

Front 27

Are the measures of central tendency (mean, median, and mode) computed using invalid responses as well as valid?

Back 27

No.

Front 28

This is the average value of all valid observations.

Back 28

Mean.

Front 29

This is the mid-point of all valid observations.

Back 29

Median.

Front 30

This is the most occuring valid observation.

Back 30

Mode.

Front 31

Where are the three measures of central tendency computed?

Back 31

In the statistics box, that can be computed alongside a frequency table.

Front 32

Are measures of central tendency helpful with categorical variables.

Back 32

No.

Front 33

How is range computed?

Back 33

Largest - smallest value

Front 34

This determines how spread out the values are. If all the cases were the same, the _______ would equal zero.

Back 34

Variance.

Front 35

What is the equation for VARIANCE?

Back 35

Sum of squared distances from the mean for all cases / number of cases - 1

(value - mean)squared + (value - mean)squared / number of cases - 1

Front 36

This is the square root of variance and is used to obtain a measure in the same units as the original data?

Back 36

Standard Deviation.

Front 37

How do you calculate standard deviation?

Back 37

Square root of variance.

Front 38

This is an expression of the standard deviation as a percentage of the mean. It allows for the comparison of variability of diferent variables.

Back 38

Coefficient of variation.

Front 39

How do you calculate coefficient of variation?

Back 39

Standard Deviation / Mean X 100

Front 40

This is used to determine the position of a case in the distribution of observed values.

Back 40

Zscore or standard score

Front 41

This tells us how many standard deviation units a case is above or below the mean.

Back 41

Zscore or standard score.

Front 42

How do you calculate Zscore?

Back 42

value - mean / standard deviation.

Front 43

If a value is in the 10th percentile, what does that mean?

Back 43

That 10% of cases are smaller than your value.

Front 44

The ____________ range is often used to show variability in data.

Back 44

Interquartile

Front 45

True or false: the interquartile range is often affected by outliers?

Back 45

False

Front 46

The 50th percentile is also referred to as what?

Back 46

The median.

Front 47

For most quantitative variables, which four characterstics tned to be of the most interest?

Back 47

1) Location (center, average - mode, median, and mean)
2) Spread (variability, range, quartiles, variance, standard deviation).
3) Shape (standard deviation)
4) Outliers

Front 48

A way to visually display measures of central tendency and dipersion (variability) in viewing quantiative variables (2 or more groups).

Back 48

Box plot

Front 49

This visually represents all individual data values

Back 49

Stem and Leaf plot

Front 50

What values do box plots and histograms both show?

Back 50

1) Median
2) Interquartile Range
3) Largest value
4) Smallest value

Front 51

T/F - the length of the whiskers is determined by the size fo the box?

Back 51

True

Front 52

How long are the whiskers on a box plot?

Back 52

up to 1.5 box length from each edge of the box.

Front 53

How far can an outlier exist visually from a box on a boxplot?

Back 53

1.5 to 3.0 box lengths from the edges of a box

Front 54

On a boxplot, how can you identify central tendency?

Back 54

The line - the median.

Front 55

How is an outlier identified on a boxplot?

Back 55

an O

Front 56

How is an extreme value identified on a boxplot?

Back 56

a *

Front 57

How far from the box is an extreme value on a boxplot?

Back 57

over 3 box lengths from the box.

Front 58

How can you identify how large your box is on a boxplot?

Back 58

It is your interquartile range - 25% - 75%

Front 59

Does the width of a box plot mean anything?

Back 59

No.

Front 60

This is similar to a histogram because it shows how many cases have various values - but differs becuase it preserves more information about values of oberservation.

Back 60

A Stem and Leaf Plot

Front 61

This looks like a histogram turned sideways.

Back 61

Stem and Leaf Plot.

Front 62

Stem width in a stem-and-leaf plot represents what?

Back 62

How far to move the decimal.

Front 63

If (Each leaf: 5) then...

Back 63

Each leaf represents 5 cases.

Front 64

This is used to numerically dispaly the comparison of the mean (averages) of one quantitative and one or more categorical variables

Back 64

Compare Means

Front 65

An explanatory variable in compare means is also called what?

Back 65

Independent Variable

Front 66

A response variable in compare means is also called what?

Back 66

Dependent Variable

Front 67

Which is dependent and independent: income and degree.

Back 67

Income: dependent
Degree: Independent

Front 68

In SPSS, is the independent variable categorical or quantitative?

Back 68

Categorical

Front 69

In SPSS, is the dependent variable categorical or quantitative?

Back 69

Quantitative

Front 70

This allows you to visually see how two diferent independent variables are related to a dependent variable (two categorical, one quantitative)

Back 70

Clustered Bar Chart

Front 71

True or False: one variable in compare means is awlays a quantitative variable?

Back 71

True

Front 72

True or False: Boxplots graphically display a quantitative variable.

Back 72

True.

Front 73

This is a table that contains counts of the number of times various combinations of values of two categorical variables occur.

Back 73

Crosstab

Front 74

Can quantitative variables be analyzed using a crosstab?

Back 74

No.

Front 75

Which variable (dependent or independent) is usually the variable on the top of a crosstab?

Back 75

Independent

Front 76

Columns are usually made up of which variable (dependent or independent)?

Back 76

Independent

Front 77

This is the best way to visually display 2 categorical variables.

Back 77

Stacked Bar Chart

Front 78

This graphically displays two quantitative variables.

Back 78

Scatterplot

Front 79

This graphically displays three or more quantitative variables.

Back 79

Scatterplot matrix

Front 80

This numerically displays two or more quantitative variables.

Back 80

Table of descriptive statistics

Front 81

Which variable is usually plotted on the x-axis on a scatterplot?

Back 81

Independent variable.

Front 82

Which variable is usually plotted on the y-axis on a scatterplot?

Back 82

Dependent variable.

Front 83

Markers in a scatterplot allow us to add what type of variable?

Back 83

Categorical.

Front 84

Can SPSS create a line in a scatterplot? If so, what is it called?

Back 84

Yes, a FIT LINE.

Front 85

This is a characteristic of a sample.

Back 85

Statistic.

Front 86

This is a characteristic of a population.

Back 86

Parameter.

Front 87

The mean of a sample is demonstrated with what character.

Back 87

_
x

Front 88

The standard deviation of a sample is demonstrated with what character.

Back 88

s

Front 89

The mean of a population is demonstrated with what character.

Back 89

u

Front 90

The standard deviation of a population is demonstrated with what character.

Back 90

o/a

Front 91

T/F = a normal distribution is symmetric and bell-shaped.

Back 91

True

Front 92

Where is the mean, median and mode on a curve of a normal distribution?

Back 92

In the middle

Front 93

How do the mean, median and mode relate to each other in a normal distribution?

Back 93

They are equal.

Front 94

Using the empirical rule, what percentage of values fall within 1 standard deviation of the mean in either direction?

Back 94

68%

Front 95

Using the empirical rule, what percentage of values fall within 2 standard deviations of the mean in either direction?

Back 95

95%

Front 96

Using the empirical rule, what percentage of values fall within 3 standard deviations of the mean in either direction?

Back 96

99.7%

Front 97

What is another name for a standardized score?

Back 97

Z-score

Front 98

This measures how many standard deviations the value (or raw score) falls above or below the mean

Back 98

Z-score

Front 99

How do you calculate Z score?

Back 99

value - mean / standard deviation

Front 100

T/F : every value has a z score?

Back 100

True

Front 101

What is the z-score for the mean of a normally distributed curve?

Back 101

0

Front 102

In a normally distributed curve, what percentage of the time does an 'unusual' event occur?

Back 102

5% of the time

Front 103

5% of the observations fall outside the range of what z-scores?

Back 103

2 and -2

Front 104

What is the p value of an unusual or statistically significant observation?

Back 104

0.05

Front 105

Using sample information to make conclusions about a broader range of individuals than just those who are observed is called what?

Back 105

Inferential statistics

Front 106

T/F: inference methods can only be applied when it is reasonable to assume that the data are representative for the question being considered about a larger group.

Back 106

T

Front 107

How does one guarantee that the sample is representative of the larger population?

Back 107

Take a random sample

Front 108

These are numerical values computed from a sample (such as mean, standard deviation, proportion of observations that fall in a particular category)

Back 108

Statistic

Front 109

These are numbers associated with a population (fixed, unchanging numbers such as "normal" body temperatures = 98.6, "average work week - 40 hours)

Back 109

Parameter

Front 110

This is defined as the distribution of all possible outcomes for a statistic (within samples of the same size)

Back 110

Sampling Distribution

Front 111

The standard deviation of the sampling distribution is called what? (for example, the standard deviation of all possible values of a sample mea)

Back 111

Standard Error

Front 112

This refers to the variability of the observations in a sample

Back 112

Standard Deviation

Front 113

This refers to the variability of a statistic

Back 113

Standard Error

Front 114

By using a histogram to evaluate a distribution of means, you would be looking at what?

Back 114

Sampling Distribution

Front 115

By computing descriptive statistics for a group of means, you would be able to calculate the standard deviations of the means which is called what?

Back 115

Standard Error

Front 116

How do you calculate the standard error of the mean?

Back 116

population standard deviation / square root of sample size in your group (o / square root of n)

Front 117

T/F: larger samples show less variability in the possible outcomes?

Back 117

True

Front 118

How do you calculate a z-score or standard score for an observed mean?

Back 118

Value - population mean / standard error of the mean

Front 119

This visually displays the interquartile range?

Back 119

Stem and Leaf Plot

Front 120

How can you visually display the distribution of z scores as it applies to the empirical rule?

Back 120

By using a Z score histogram

Front 121

Best option for numerically displaying the number of cases who gave each response for ONE variable.

Back 121

Frequency Table

Front 122

Used for graphically displaying the number of cases who gave each response for ONE variable.

Back 122

Pie Chart
Simple Bar Chart
Histogram
Box Plot
Stem and Leaf Plot

Front 123

Best option for numerically viewing the averages of ONE quantitative variable across categories of ONE qualitative variable

Back 123

Compare Means

Front 124

Best choice for graphically displaying the averages of ONE quantitative variable across categories of ONE qualitative variable.

Back 124

Simple Bar Chart

Front 125

Best option for numerically viewing the averages of ONE quantitative variable across categories of TWO qualitative variables.

Back 125

Compare Means

Front 126

Best choice for graphically displaying the averages of ONE quantitative variable across categories of TWO qualitative variables.

Back 126

Clustered Bar Chart

Front 127

Best option for numerically viewing the relationship between two categorical variables. In other words, this shows the counts of cases in each response category of one categorical variable across response categories of another categorical variable.

Back 127

Crosstab

Front 128

Best choice for graphically displaying the relationship between two categorical variables.

Back 128

Stacked Bar Chart

Front 129

Best option for numerically viewing the relationship between two quantitative variables?

Back 129

Tables of descriptive statistics

Front 130

Best option for graphically viewing the relationship between two quantitative variables?

Back 130

Scatterplot