Week 3 - Measurement concepts and foundations of quantitative research/data

Apart of Introduction to Evidence Based Practice and Research in Health Sciences at UniSA

What is the difference between reliability and validity?

Reliability is the consistency or repeatability of measurements whereas validity is the accuracy of measurements (of what was supposed to measure).

What are the three types of reliability?

• Intra-rater reliability

• Inter-rater reliability

• Test-retest reliability

Intra-rater reliability

Reliability where the same measurement undertaken by the same person at different time points.

Inter-rater reliability

Reliability where the same measurement undertaken by different people at the same time point.

Test-retest reliability

Reliability where the same measurement undertaken at different time points, usually used with tools or instruments.

What are the four types of validity?

• Face validity

• Content validity

• Criterion validity (predictive and concurrent)

• Construct validity

Face validity

Validity where the measurement appears to measure what it aims to measure. (e.g. weight scales)

Content validity

Validity where the measurement covers everything it should. (e.g. visual analogue scale which only measures pain severity)

What is the difference predictive and concurrent criterion validity?

Although both relate to an outcome, predictive is the ability to predict future occurrences whereas concurrent is the ability to correlate other measures that have already been validated.

Construct validity

Validity with how well the measure truly reflects what it claims to measure (e.g. measuring joint swelling using a tape measure).

Normal distribution aka bell-shaped curve

A naturally occurring phenomena which describes how the values of a variable are distributed. It's symmetrically distributed where most of the observations cluster around the central peak.

How is population and sample represented in quantitative research?

The population is represented in Greek letters, where the size is represented as 'N' and a sample which is drawn from the whole population is represented in English letters, where the size is represented as 'n'.

Variable

Any characteristic, number, or quantity that can be measured or counted like age, gender, country of birth, type of program, height, weight etc…

Data

Factual information (either numerical or narrative) that are collected where conclusions may be drawn.

What is the difference between qualitative nominal and ordinal data?

Nominal data can be categorised like hair colour whereas ordinal data can be ranked like letter grades.

What is the difference between quantitative discrete and continuous data?

Discrete data can be counted like the number of students in a class whereas continuous data is measured on a scale or continuum like the height of children.

What are three common ways categorical data is visually represented?

Frequency table, bar chart/graph, and pie chart.

What are four common ways numerical data is visually represented?

Frequency distribution table, histogram, box plots aka box and whisker diagrams, and line graphs.

What to look for when summarising descriptive statistics (3 points)?

• Summary measures

• Measures of central tendency

• Measures of dispersion

What are the three measures of central tendency?

• Mode

• Mean

• Median

Mode and what are the limitations?

• A measure of central tendency which is the most commonly occurring value in a data set/distribution, which can be used for both numerical and categorical data.

• It has its limitations where it may not reflect the centre of distribution and can be more than one mode or none at all.

Mean and what symbol represents population and sample?

• A measure of central tendency which is the sum of the value of each observation in a quantitative dataset divided by the number of observation, which is influenced by outliers.

• The population mean is represented by μ whereas sample is x̄.

Median

A measure of central tendency which is the middle value in distribution when the values are arranged in ascending or descending order, which is not affected by outliers.

What is the difference between symmetric and asymmetric/skewed distribution?

Symmetric is where all measures of central tendency are in the middle of the distribution whereas asymmetric is where the mode and median are normal however, the mean is generally pulled in the direction of the tails, which can be positively/right skewed or negative/left skewed.

Outliers

Sample values that lie far away from the vast majority of other sample values, which occur due to error in data capture or entry in which case outlier can be removed however if it's accurate, it is included.

What are the four measures of dispersion?

• Range

• Standard deviation

• Quartiles

• Interquartile range

Range

A measure of dispersion which is a basic measure of data dispersion from largest to smallest value in a data set.

Standard deviation

A measure of dispersion which is an average deviation of observation from the mean, where large implies that data is widely spread and small implies data is mainly concentrated around the mean.

Quartiles

A measure of dispersion where data divided in the four equal parts, denoted by quartile 1 (Q1), quartile 2 aka median (Q2), and quartile 3 (Q3).

Interquartile range

A measure of dispersion which measures the range of the middle 50% of the values to determine the difference between the upper and lower quartiles using Q3-Q1.