Chapter 3 Statistical Essentials II

Imagine you are to conduct a study on how weight and age group (18-35, 36-53, and over 54) relate to systolic blood pressure. What are the variables in this study? Characterize each variables in terms of discrete vs. continuous. Qualitative vs quantitative. Independent vs dependent. and level of measurement.

1. Age: continuous quantitative independent ordinal 2. weight continuous quantitative independent ratio 3. systolic bp continuous quantitative dependent ratio

Three correlation coefficients

1. Cronbach's alpha 2. Kudar-Richardson's coefiicient (k-20) 3. Cohen's Kappa

Variables

A trait or characteristic whose value is not fixed and can change (either from subject to subject, or within the same subject over time).

True or False: The length of time in minutes that a patient waits to be called at a hospital is an interval level of measurement

False (this is a duration of time-- it's interval at the clock times)

True or False: Level of internal validity is independent of the presence of confounding variables.

False.

True or False: Validity always precedes reliability: An instrument can be valid but not reliable.

False. Reliability precedes validity. An instrument can be reliable but no valid. To be valid you must be both reliable and valid. To make inferences about statistical tests you must have both.

True or false: A discrete variable can have any possible value in a defined continuum.

False. a continuous variable would have any possible value in a defined continuum.

What is the difference between an independent variable and a dependent variable? Give examples of each.

Independent: the manipulated object by researcher (effects other variables but other variables do not effect it.) Dependent variable: variable that changes depending on other variables. example: years exercising on muscle tone.

Kudar-Richardson Coefficient (k-20)

Internal consistency-- used with nominal/ordinal level of measurement

criterion-related validity

The degree to which measurements from one tool or instrument may be correlated with measurements from other valid and reliable instruments.

Validity

The extent to which a test measures the variable it is designed to measure

Levels of measurement definition

The four different scales of measurement, used to differentiate types of data (and the statistical procedures appropriate for the data).

External validity

The level of confidence in whether or not the results of a study may be generalized from the sample to the target population.

Reliability

The measure of whether or not a test is able to consistently measure a given variable (not always accurate)

True or False: A study that can be used to make accurate inferences about its sample population can be said to have high external validity.

True.

Cohen's Kappa

Used with Interrater Reliability (used to determine degree of agreement between individuals' scores on ratings). --- Uses scale of 0 to 1 with 1 indicating complete agreement

The coefficient used to measure reliability of interval or ratio measurements is: a. Kuder-Richardson (K-20) coefficient b. Cronbach's alpha c. Cronbach's kappa. d. None of these are correct.

b. cronbach's alpha **Within internal consistency. cronbach's alpha (0 to 1, 1 being most consistent). Kudar-Richardson (K-20) is ordinal or nominal

Which of the following statement accurately desribes measurement error in research? a. Incorrect use of instruments leads to random errors. b. measurement error is universal. c. random errors occur consistently d. systematic errors result from unknown causes.

b. measurement error is universal.

Which of the following is the example of a continuous variable? a. zip code b. gender c. income d. profit vs non profit nursing home

c. Income. (at ratio level and therefore continuous. Remainder are nominal)

data set

collection of data values

Transforming a variable's level of measurement from a higher level to a lower level always results in: a. increased measurement error. b. categorical variables. c. continuous variables d. loss of information

d. loss of information

True or false: an instrument can be valid without being reliable

false.

content validity

if a tool measures all aspects of the idea of interest/ the degree which a measurement tool captures the elements of the concept of interest

Interrater Reliability

the ability of a test or scale to provide consistent values when used by different people

Hawthorne Effect

the alteration of behavior by the subjects of a study due to their awareness of being observed.

construct validity

the degree to which an instrument or tool measures the specific idea of interest ( is this depression test only testing depression or anxiety too?)

Internal validity

the degree to which changes on the dependent variable may be attributed to the independent variable (as opposed to the confounding variables). Strongly influenced by the quality of the study and control of confounding variables.

Measurement error

the difference between a true value of a variable and the value that has been measured.

Test-retest reliability

the measure of a test's ability to consistently provide the same measurements across time

test-retest reliability

the measure of a test's ability to consistently provide the same measurements across time. Taking and retaking tests at different times

independent variable

the variable that influences another variable(s). Also, the variable that the investigator controls or manipulates to affect the dependent variable

Instrument

tool/device for measuring variables

True or false: An investigator asks a patient participant at a local hospital to rate the service as outstanding, fair, poor, or very poor. These data are measured on an ordinal level of measurement

true

data

values of variables when they change (ie: bp= variable, each measurement=data)

Qualitative variables

variable whose data values are nonnumeric

Quantitative variable

variables whose data values are numeric

3 types of validity

1. content 2. criterion-related 3. construct

Examples of Nominal level of measurement

1. gender 2.ethnicity 3. religious affiliation 4. political affiliation 5. hair color 6. treatment results (improvement or recurrence) 7. signs and symptoms (present or not present)

Examples of Ordinal level of measurements

1. grouped age (18 & younger, 29-30, etc) 2. letter grade 3. likert scale 4. ranking in a race (first, second, third) 5. histological ratings (+,++,+++) 6. pain scale (0-10) 7. patient's satisfaction 8. Nursing performance

Examples of Ratio level of measurement

1. income 2. age 3. height 4. weight 5. blood pressure 6. amount of time

3 Statistical evaluations of Reliability

1. internal consistency 2. test-retest 3. interrater reliability

Factors that Influence instrument reliability

1. length of tool (shorter less reliable b/c missing aspects) 2. Clarity of expression of each item (are questions confusing) 3. Length of time allowed for measurement tool completion 4. Condition of test takers (are they sick) 5. Level of difficulty of tool (is it too easy or difficult for test-takers) 6. Homogeneity of subjects (similar subjects likely to answer similarly)

Four types of levels of measurement

1. nominal 2. ordinal 3. interval 4. ratio

Interval level of measurement examples

1. temperature 2. academic standardized tests/iq tests 3. depression score 4. time of day 5. dates

Categorical variables

A variable that has a finite and fixed number of possible values ( ie, every value is assigned to a particular group or category); variables measured at the nominal and ordinal levels

Discrete variable

Also known as categorical variable; a variable that can be counted, but has a finite number of countable categories; variables measured at the nominal and ordinal level of measurement (numeric, countable, not a fraction)

Confounding variables

Any uncontrolled variable that may influence the outcome of a study

How strength of internal validity is often evaluated

Based on whether there is uncontrolled or confounding variables

Interval level of Measurement

Data are classified into categories with rankings and are mutually exclusive as in the ordinal level of measurement. In addition, specific meanings are applied to the distances between categories. These distances are assumed to be equal and can be measured. There is no absolute value of 0 (0 degrees F does not mean there is no temp). There's no concept of ratio/equal proportion (25F is not 3x as cold as 75F)

Ordinal Level of Measurement

Data are classified into mutually exclusive categories; and, ranking or ordering is imposed on categories. Cannot tell the distance between two categories (ie, amount of pain between 1 and 2 on pain scale may be difference between the amount of pain between 3 and 4)

Nominal Level of Measurement

Data classified into mutually exclusive categories where no ranking or ordering is imposed on categories. (name or category)

Random Error

Errors that occur by chance and are the result of unknown causes

Systematic Error

Errors that occur consistently because of known causes. One common source of systematic error is the incorrect use of tools or instruments

Cronbach's alpha

Within internal consistency measuring if items within a tool measure the same thing-- this coefficient used on interval-ratio level of measurement

internal consistency reliability

a measure of whether or not items in the same test, which purport to measure the same variable, are related. (are they consistent with one another?)

continuous variables

a variable that has an infinite number of possible values ( ie, every value on a continuum) or the infinite number of values between two consecutive values; variables measured at the interval and ratio level. (a range)

Scientists are studying the effects of a drug on cancer. Which of the following would be considered an independent variable in the study? a. timing of the doses b. effect of drug on cancer c. side effects of the drug d. all of the above

a. timing of the doses

Ratio Level of Measurement

all characteristics of the interval level of measurement are present; in addition, there is a meaningful zero, and ratio or equal proportion is present.

Construct

an idea or concept of interest

dependent variables

an outcome-variable that is affected or influenced by the independent variable