Quiz 3: Variables and Measurements

what are variables?

characteristics of individuals, objects, or environmental conditions that may have more than one value, anything in research that can change or vary

what different kinds of variables may exist in research?

independent variables, dependent variables, and extraneous variables, also discrete or continuous

what are independent variables?

Variables that are manipulated or changed in an experiment to test their effect on dependent variables in intervention studies. may also be used in predictive/prognositic studies as the presumed predictor of an outcome of interest

what is a dependent variable?

The variable that is measured or observed in an experiment, which is influenced by changes in independent variables. It represents the outcome or effect being studied.

what are extraneous variables?

confounders, confounds the relationship between the independent and depedent variables, anything that could interfere, need to be anticipated and controlled for if possible

what are discrete variables?

Variables that have distinct, separate values, often counted in whole numbers rather than measured. dichotomous or polytomous

what are continuous variables?

Variables that can take on any value within a given range, including fractions and decimals. They are measured rather than counted and can represent an infinite number of values. (theoretically can assume infinitely finer degrees of measurement depending upon instrument used)

what is measurement?

a method of assigning descriptive or numeric values to variables, have clearly defined and consistently applied rules, used to gain understanding about the roles and behaviors of variables in a study, required step conducting using statistics analysis

how is measurement classified?

classification of values into 4 different types- nominal. ordinal, interval, and ratio (not always straightforward)

what are nominal measurements?

classification without value placed on category, no ranking or order, can use names or numbers that are symbols, not quantities (ex: sex, race/ethnicity/religious affiliation)

what are ordinal measurements?

Classification with a rank order but without equal intervals between levels, can use numbers to label categories, but can’t do math bc they are symbols and not quantities (ex WB status, level of assistance required, MMT grade)

what are internal measurements?

Classification with rank order and equal intervals between levels, allowing meaningful comparisons of differences. the origin is unknown (no zero point), positive and negative values possible, can do addition and subtraction. (Examples include temperature measured in Celsius or Fahrenheit, calendar year)

what are ratio measurements?

Classification with rank order, equal intervals, and a true zero point, allowing for a full range of mathematical operations including addition and subtraction and multiplication and division. (Examples include weight, height, circumference, blood pressure, speed, distance)

what are reference standards?

research measurements are compared to know measures/findings

what are gold standard standards vs norm referenced standards vs criterion references standards?

Gold standard standards are the highest level of measurement accuracy, often used as a benchmark and are considered the most reliable and accurate method, while norm-referenced standards compare an individual's performance to that of a larger and comparable group. Criterion-referenced standards evaluate an individual's performance against a previously established absolute standard

what are measures a combination of?

true value and error (noise- which is the difference between the actual measure and the true measure)

what can errors come from?

instrument error, observer error, and patient variability

what is meant by responsiveness to change?

the ability of a measure (instrument) to detect the change in the phenomenon of interest, depends on the fit between the instrument and the operational definition of a variable, the number of values on the measurement scale, and the standard error of measurement

what are the floor and ceiling effects?

These refer to limitations in a measurement instrument where a score cannot be accurately represented due to the instrument being unable to detect higher (ceiling effects) or lower values (floor effects) than the limits of its scale, potentially masking true changes in the variable being measured.

how can change be interpreted?

as meaningful change (change that makes an impact, that matters), can be represented as the MCID (minimally clinically important difference)w

what is MCID?

MCID, or minimally clinically important difference, is the smallest change in a patient's condition that is deemed significant enough to be noticeable by the patient or to warrant a change in treatment. It serves as a threshold to interpret whether a change in a measurement is meaningful.

what is reliability vs validity?

Reliability refers to the consistency of a measurement instrument, indicating that it produces stable and repeatable results over time. Validity, on the other hand, assesses whether the instrument measures what it is intended to measure, reflecting the accuracy and relevance of the results. Reliability is a necessary but not sufficient condition for validity

what are sources of error for reliability?

errors made by the examiner, subject variability, and instrumentation flaws or failures

what are the types of measurement reliability for the instrument?

test-retest, internal consistency, parallel forms, and split-half

what are types of measurement reliability for the rater?

intra-tester (within), and inter-tester (between or among)

describe test-retest reliability

A method to assess the consistency of a measure across time by administering the same test to the same group on two or more occasions and comparing the results. the time period must be short enough to avoid natural change, but long enough to avoid subject fatigue on retest

describe instrument consistency reliability

Instrument reliability refers to the degree to which an instrument consistently measures a construct they are intended to define and not with others (ex: patient questionarres)

describe parallel form reliability

A method of assessing the consistency of measurements across different version of a test that measure the same construct. This involves administering two equivalent forms of the test to the same subjects and comparing the results. purpose is to establish equivalency of 2 instruments

describe split half reliability

A method of measuring reliability that involves dividing a single test into two equal halves and assessing the consistency of scores between them. It can be used to evaluate internal consistency reliability (given a high level of agreement). is useful when can’t do multiple testing sessions to evaluate reliability

describe intra-rater reliability

Intrarater reliability refers to the degree of agreement or consistency in measurements or assessments made by the same individual across multiple instances (at least 2 times). usually with a small amount of time between the tests to reduce practice effects

describe inter-rater reliability

Interrater reliability refers to the degree of agreement or consistency between measurements or assessments made by 2 different individuals (raters) evaluating the same phenomenon. It is crucial for ensuring that the results are not dependent on who is conducting the assessment. there is no need for a gap in time, but typically blind the raters to each others’ tests and measures

what is relative reliability?

Relative reliability refers to the consistency of measurement ability to rank upon repetition among individuals, allowing for comparison of performance relative to a standard. It determines how well scores maintain their rank order over time or between different observers.

what is absolute reliability?

Absolute reliability refers to the extent to which repeated measurements or assessments yield the same results within the same individual or condition. It focuses on the degree of variation around an individual's scores, providing a clear understanding of the measurement's stability and precision.