measurement and descriptive analysis Exam 1

representation of data sets. describes the study participants

descriptive statistics

application of statistical methods to the data sets. descriptive results are shown and the results of the participants are shown.

inferential statistics

pvalue < 0.05

significant from the empirical rule

displayed on the x axis

independent variable

displayed on the y axis

dependent variable

what are examples of discrete/categorical variables?

nominal and ordinal

What is nominal data?

Data that can be placed in categories (colours, Type A/B personality, types of attachment style).

what is ordinal data?

Data that can be placed in a sequence (e.g. best, worst, okayish; oldest - youngest, 1-10 scale).

what are continuous variables?

number line. interval and ratio

contains no true 0. only + and - numbers

interval

contains - and only includes + numbers

ratio

what is the empirical rule?

68-95-99.7

68% of all data fall within __ SD of the mean

1

95% of the data fall within __ SD of the mean

2

99% of the data fall within __ SD of the mean

3

68% of the empirical rule represents what percentiles?

16-84

95% of the empirical rule represents what percentiles?

2.5 to 97.5

99% of the empirical rule represents what percentiles?

0.5-99.5

how do you organize and visualize the box and wisker plot?

...

how does a histogram look?

bar chart with the but the bars touching. the mean, median, mode are the same

if RR is less than 1, what does it mean?

we have less of a risk

if RR greater than 1, what does it mean?

we have a higher risk

if RR is =1, what does that mean?

no change

how do you calculate risk?

number of individuals with unfavorable outcome/total number

how do you calculate relative risk?

risk in treatment group/risk in the control group

how much less of a likelihood there is in one group relative to another

relative risk reduction

how do you calculate RRR?

1-RR(intervention group)

how do you calculate absolute risk reduction (ARR)?

% risk in control group - %risk intervention group

how do you calculate number needed to treat?

1/ARR

what is considered a good NNT?

NNT<20

Do you round up or down for NNT?

up

how is NNH calculated?

same as NNT but for a negative effect; 1/ARR.

do you round up or down for NNH?

down

how do you calculate odds ratio?

ad/bc

how do you set up the odds ratio table?

exposureno exposureoutcomeno outcomea,b,c,d

how do you calculate hazard ratio?

hazard rate in tx group / hazard rate in control group

what does it mean if the odds or hazard ratio is 1:1?

non factor

what does it mean if the odds or hazard ratio is >1:1?

risk factor

what does it mean if the odds or hazard ratio is <1:1?

protective factor

how do you set up a truth table?

test positivetest negativedisease no diseasetrue positive, false positive, false negative, true negative

people who have the disease

sensitivity calculation

people who dont have the disease

specificity calculation

how do you calculate sensitivity?

TP/(TP+FN)

how do you calculate specificty?

TN/(TN+FP)

how do you calculate positive predictive value?

TP/(TP+FP)

how do you calculate negative predictive value?

TN/(TN+FN)

probability that a finding will be present in the diseased patient/probability that it will be present in the non-diseased patient

Likelihood Ratio+

the probability that a finding will not be present in the non diseased patient/probability that it will not be present in the diseased patient

LR-

how do you calculate LR-?

(100-sensitivity/specificity)

how do you calculate LR+?

sensitivity/(100-specificity)

LR>10

Strong evidence to rule in disease

LR 5-10

Moderate evidence to rule in disease

LR 2-5

Weak evidence to rule in disease

LR 0.5-2

no change

LR 0.2-0.5

Weak evidence to rule out disease

LR 0.1-0.2

Moderate evidence to rule out disease

LR<0.1

Strong evidence to rule out disease