Quantitative - data analysis issues: levels of measurement, error types, descriptive

Question 1

What are the data in quantitative research?

Answer 1

primarily numerical

descriptive statistics - ways of displaying data and summarising it in ways that are easily understood

Question 2

In what ways are numbers used to display data?

Answer 2

• numerical result (e.g.BP, age)
• coded category (e.g. 1 = male 2= female)
• ordered categories (e.g. pain scale)

Question 3

What are the levels of measurement?

Answer 3

nominal, ordinal, interval, ratio

Question 4

What are the properties of nominal data?

Answer 4

-different categories

Question 5

What are the properties of ordinal data?

Answer 5

• different categories

- categories can be ranked

Question 6

What are the properties of interval data?

Answer 6

• different categories
• categories can be ranked
• equal distances between categories

Question 7

What are the properties of ratio data?

Answer 7

• different categories
• categories can be ranked
• equal distances between categories
• fixed zero

Question 8

What are the ways of presenting descriptive data?

Answer 8

• tables: allows data from different variables to be displayed togethe
• charts: immediate visual impact
• measures of central tendancy: mean, median, mode
• measures of dispersion: range, interquartile range, standard deviation, variance

Question 9

Why do we perform statistical analysis?

Answer 9

to draw inferences from the sample that we studied about the population of interest

Question 10

What are the two basic approaches to statistical analysis?

Answer 10

• hypothesis testing (using P values)

- estimation (using confidence intervals)

Question 11

How does hypothesis testing happen?

Answer 11

-set null hypothesis, set study hypothesis, carry out significance test, obtain test statistic, compare test statistic to hypothesised critical value, obtain P value, make decision

Question 12

What is a P value?

Answer 12

P value = PROBABILITY of obtaining the study results in the Ho is true

• can be between 0 and 1
• the closer it is to 0, the more likely it is that the Ho should be rejected
• statistical sig - often set at 5
• only tells you how likely the results are when the Ho is true

Question 13

How do you know if there is sufficient or insufficient evidence to reject or accept the Ho?

Answer 13

if P > or = 0.05 there is insufficient evidence to reject the Ho

Question 14

What is a Type I error?

Answer 14

(false positive error)

incorrect rejection of a true null hypothesis

Question 15

What is a Type II error?

Answer 15

(false negative error)

failure to reject a false null hypothesis

Question 16

What is the power of a study?

Answer 16

the probability of being able to detect a difference between the study groups, should one exist

• usually expressed as a %
  e. g. - for a study with 80% power theres an 80% chance of detecting a real difference between study groups

Question 17

When is a confidence interval calculated?

Answer 17

when info about the effect size and whether the results are of clinical significance

-measure of the precision (accuracy) with which the quantity of interest is estimated

Question 18

All you know about confidence intervals

Answer 18

• calcualted from any estimated quantity from the sample data such as mean and median etc
• 95% CI is the range of values within which the true population quantity would fall 95% of the time if the study were repeated multiple times
• 95% confident that the ‘true’ value lies within the specified range
• if range includes 0 then may be no difference between groups