Quantitative - data analysis issues: levels of measurement, error types, descriptive Flashcards
What are the data in quantitative research?
primarily numerical
descriptive statistics - ways of displaying data and summarising it in ways that are easily understood
In what ways are numbers used to display data?
- numerical result (e.g.BP, age)
- coded category (e.g. 1 = male 2= female)
- ordered categories (e.g. pain scale)
What are the levels of measurement?
nominal, ordinal, interval, ratio
What are the properties of nominal data?
-different categories
What are the properties of ordinal data?
- different categories
- categories can be ranked
What are the properties of interval data?
- different categories
- categories can be ranked
- equal distances between categories
What are the properties of ratio data?
- different categories
- categories can be ranked
- equal distances between categories
- fixed zero
What are the ways of presenting descriptive data?
- 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
Why do we perform statistical analysis?
to draw inferences from the sample that we studied about the population of interest
What are the two basic approaches to statistical analysis?
- hypothesis testing (using P values)
- estimation (using confidence intervals)
How does hypothesis testing happen?
-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
What is a P value?
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
How do you know if there is sufficient or insufficient evidence to reject or accept the Ho?
if P > or = 0.05 there is insufficient evidence to reject the Ho
What is a Type I error?
(false positive error)
incorrect rejection of a true null hypothesis
What is a Type II error?
(false negative error)
failure to reject a false null hypothesis
What is the power of a study?
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
When is a confidence interval calculated?
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
All you know about confidence intervals
- 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