Introduction to Inferential

statistics & Important

Statistical Tests

Dr. Adeel Ahmed Khan

MBBS, FCPS (Community Medicine)
Public Health Specialist
Ministry of Health, KSA
Learning objectives
 Concepts of hypothesis testing
 Point interval & Confidence intervals
 T-tests
 Choosing the right test
 What is a non-parametric test?
 Statistics scenarios
 Statistics

Descriptive Inferential

Hypothesis
Estimation
testing

Point
estimates

Interval
estimates (CI)
Hypothesis Testing
 Decision making process for evaluating claims about
a population

 Researcher must define

 Population under study
 State particular hypothesis
 Give significance level
 Select a sample from the population
 Collect the data
 Calculate statistical test
 Conclusion
Methods to test hypothesis

 The Traditional method

 The Confidence Interval method

 The P-value method

Traditional method of Hypothesis
Testing
 Statistical hypothesis
 Is a assumption about a population parameter, which may be or
may not be correct

 Null hypothesis - Ho
 States that there is no difference between a parameter and a specific
value
 The mean of a population , equals the hypothesized value o

 Alternative hypothesis - H1
 States the existence of a difference between a parameter and a
specific value
 The mean of a population , is not equals the hypothesized value o
Steps to undertaking a Hypothesis test

Define study question

Set null and alternative hypothesis Choose a

suitable
Calculate a test statistic test

Calculate a p-value

Make a decision and interpret

your conclusions
T-tests
Paired or Independent (Unpaired) Data?

T-tests are used to compare two population means

₋ Paired data: same individuals studied at two
different times or under two conditions PAIRED
T-TEST
₋ Independent: data collected from two separate
groups INDEPENDENT SAMPLES T-TEST
Comparison of hours worked in 1988 to
today
Paired or unpaired?
If the same people have reported their hours for 1988 and
2014 have PAIRED measurements of the same variable
(hours)
Paired Null hypothesis: The mean of the paired differences
=0

If different people are used in 1988 and 2014 have

independent measurements
Independent Null hypothesis: The mean hours worked in
1988 is equal to the mean for 2014
H 0 : 1988   2014
 SPSS data entry
Paired Data

Independent Groups
Example: Triglycerides

 In a weight loss study, Triglyceride levels were

measured at baseline and again after 8 weeks
of taking a new weight loss treatment.
 Example: t-Test Results
 
95% Confidence
Interval of the  
   
Difference  
  Std. Std. Error    
Mean Sig. (2-
Deviation Mean t df
tailed)
Lower Upper

Triglyceride level at
week 8 (mg/dl) -
-11.371 80.360 13.583 -38.976 16.233 -.837 34 .408
Triglyceride level at
baseline (mg/dl)
Null Hypothesis is:

P-value =
Decision (circle correct answer): Reject Null/ Fail to reject
Null

Conclusion:
 Example: Solution
 
95% Confidence
Interval of the  
   
Difference  
  Std. Std. Error    
Mean Sig. (2-
Deviation Mean t df
tailed)
Lower Upper

Triglyceride level at
week 8 (mg/dl) -
-11.371 80.360 13.583 -38.976 16.233 -.837 34 .408
Triglyceride level at
baseline (mg/dl)

As p > 0.05, FAIL to reject the

null
P(t< -0.837) P(t>0.837)
NO evidence of a difference in =0.204 =0.204
the mean triglyceride before
and after treatment
-0.837 0.837
Example: Weight Loss

 Weight loss was measured after taking either

a new weight loss treatment or placebo for 8
weeks

Treatment Std.
N Mean
group Deviation
Placebo 19 -1.36 2.148
New drug 18 -5.01 2.722
 Example: t-Test Results
Levene's Test
95% CI of the
for Equality of T-test results
Difference
Variances
Sig. Mean Std. Error
F Sig. t df Lower Upper
(2-tailed) Difference Difference
Equal variances
2.328 .136 4.539 35 .000 3.648 .804 2.016 5.280
assumed
Equal variances
4.510 32.342 .000 3.648 .809 2.001 5.295
not assumed

Null Hypothesis is:

P-value =
Decision (circle correct answer): Reject Null/ Fail to reject
Null
Conclusion:
 Example 2: Solution
Levene's Test
95% CI of the
for Equality of T-test results
Difference
Variances
Sig. Mean Std. Error
F Sig. t df Lower Upper
(2-tailed) Difference Difference
Equal variances
2.328 .136 4.539 35 .000 3.648 .804 2.016 5.280
assumed
Equal variances
4.510 32.342 .000 3.648 .809 2.001 5.295
not assumed

H0: μnew = μplacebo As p < 0.05, DO reject the null

IS evidence of a difference
P(t< -4.539) P(t>4.539)
in weight loss between Is < 0.001 Is < 0.001
treatment and placebo

-4.539 4.539
Uses of t-test
 To compare mean of a single sample to a population mean
or a hypothetical mean

 Population standard deviation is unknown

 Underlying variable is normally distributed

 Paired t-test

 Appropriate test to compare means on the same and

related samples over a time or in different situations

 Provides a hypothesis test of difference between

population means for a pair of random samples

 Can be used as a before –after method or a repeated

measures method

 Also called as dependent t-test

Example

An investigator wishes to test the effect of a treatment

on the well-being of depressed individuals; a score that
could range from 0 to 20, higher scores indicate greater
well-being (that is , treatment is having a positive effect).

The scores of nine individuals were recorded

before and after receiving treatment.

Use α=5% to test whether mean scores before and after

receiving treatment is different or not
 Example

 A diet claims to cause more than a 10 lb weight loss over

a six months period, a researcher designed a test using
16 individuals before and after weight. The samples are
paired by each individual.

 Mean (differences) = 10.56

 SD (Differences) = 11.56

a) Researcher wants to know is the diet truly delivers

greater than10 lb?
 Dependent or independent sample….

 A researcher was interested whether treatment A is

better than treatment B for abscess reduction? So he
randomly divided pts. in two groups. Gave treatment A to
first group and B to the other. Size of abscess was obs.

 The researcher was interested in effectiveness of a

particular diet in slimming center. So wt of clients before
and after was obs.
 Statistics

Descriptive
Estimation
 Estimation: the process by which the sample
data is used to indicate the value of an
unknown quantity in the population:
 Point estimate
 Confidence intervals
Types of estimators

 Point estimator gives a single value as an

estimate of the parameter of interest

 Interval estimator specifies a range of values

of the parameter and our confidence that the
parameter value is in that range
 Confidence Intervals

 A range of values within which we are confident (in terms of

probability) that the true value of a pop parameter lies
 A 95% CI is interpreted as 95% of the time the CI would
contain the true value of the pop parameter
 i.e. 5% of the time the CI would fail to contain the true value
of the pop parameter
 Exercise
 Discuss what the interpretation is for the confidence interval from
this Example (Weight loss was measured after taking either a new
weight loss treatment or placebo for 8 weeks) highlighted below:
 Large-Sample Confidence Interval for 
x  z 
n
2

• where Z/2 is the Z value with an area /2 to its

right

• The parameter  is the standard deviation of the

population from which sample is drawn and n is the
sample size.

• When  is unknown (most cases) and n is large, the

value of  can be approximated by the sample
standard deviation (s).
 Two categorical variables
Are boys more likely to prefer maths and science than girls?

Variables:
 Favourite subject (Nominal)
 Gender (Binary/ Nominal)

Summarise using %’s/ stacked or multiple bar charts

Test: Chi-squared
Tests for a relationship between two categorical variables
Purpose of Chi square test
 To present categorical data
 Chi-square is used when the data is in
qualitative form (categories) and you want to
find out statistically significant difference
between the two or more categories
 E.g. whether pts are normotensive, hypertensive,
hypotensive; male, female; smoker, non-smoker
Basics of a Chi-square test
 For a given phenomenon, Chi-square
compares the given (observed) frequencies
with the expected frequencies

Contracted Did not

malaria Contract
malaria Total
Chloroquine 19 587 606
given
Chloroquine 193 2741 2934
not given
212 3328 3540
Example
 A researcher is interested in finding out
whether drinking and smoking are dependent
upon each other? Test the hypothesis using α
= 0.05

Smokin None Low Medium Heavy Total

g status
Smoker 1880 2048 194 76 4198
s
Non- 4290 2430 172 37 6929
smoker
s
6170 4478 366 113 11,127
  At 5% level of significance, test the
hypothesis that whether the prevalence of
alcohol consumption is differs between
different ethnicities

Ethnicities None Low Medium Heavy Total

Black 411 253 12 5 681
Hispanic 1459 757 53 10 2279
Caucasian 3732 3179 284 90 7285
Other 322 187 10 4 523
Total 5924 4376 359 109 10768
Choosing the right test
Choosing the right test
 One of the most common queries in stats support
is ‘Which analysis should I use’

 There are several steps to help the student decide

 When a student is explaining their project, these

are the questions you need answers for
Choosing the right test
1) A clearly defined research question

2) What is the dependent variable and what type of variable is it?

3) How many independent variables are there and what data

types are they?

4) Are you interested in comparing means or investigating

relationships?

5) Do you have repeated measurements of the same variable for

each subject?
Research question
 Clear questions with measurable quantities

 Which variables will help answer these

questions

 Think about what test is needed before

carrying out a study so that the right type of
variables are collected
Dependent variables

INDEPENDENT DEPENDENT
(explanatory/ (outcome)
affects
predictor) variable
variable

Does attendance have an association with exam

score?
Do women do more housework than men?
What variable type is the dependent?

Dependent

Continuous Categorical
Are boys better at maths?

 How can ‘better’ be measured and what type

of variable is it?
Exam score (Scale)

 Do boys think they are better at maths??

I consider myself to be good at maths (ordinal)
How many variables are involved?
 Two – interested in the relationship

 One dependent and one independent

 One dependent and several independent variables: some

may be controls
Data types

Research question Dependent/ Independent/

outcome variable explanatory variable
Does attendance have an Exam score (scale) Attendance (Scale)
association with exam score?
Do women do more Hours of Gender (binary)
housework than men? housework per
week (Scale)
Exercise:
How would you investigate the following topics? State the
dependent and independent variables and their variable
types.
Research question Dependent/ Independent/
outcome variable explanatory variable
Were Americans more likely to
survive on board the Titanic?

Does weekly hours of work

influence the amount of time
spent on housework?
Which of 3 diets is best for
losing weight?
Exercise: Solution
How would you investigate the following topics? State the
:dependent and independent variables and their variable
types.
Research question Dependent/ Independent/
outcome variable explanatory variable
Were Americans more likely to Survival (Binary) Nationality (Nominal)
survive on board the Titanic?

Does weekly hours of work Hours of housework Hours of work (Scale)

(Scale)
influence the amount of time
spent on housework?
Which of 3 diets is best for Weight lost on diet Diet (Nominal)
losing weight? (Scale)
Comparing means
 Dependent = Scale
 Independent = Categorical

 How many means are you comparing?

 Do you have independent groups or repeated

measurements on each person?
Comparing means
Independent
t-test
2
Comparing BETWEEN
groups
3+

Comparing
means
Comparing
2 Paired t-test
measurements
WITHIN the same
subject

3+
Comparing means
Independent
t-test
2
Comparing BETWEEN
groups One way
3+ ANOVA

Comparing
means
Comparing
2 Paired t-test
measurements
WITHIN the same
subject

3+ Repeated
measures
ANOVA
ANOVA = Analysis of variance
Exercise – Comparing means
Research question Dependent Independent Test
variable variable
Do women do more Housework (hrs Gender
housework than men? per week) (Nominal)
(Scale)
Does Margarine X Cholesterol Occasion
reduce cholesterol? (Scale) (Nominal)
Everyone has
cholesterol measured
on 3 occasions

Which of 3 diets is best Weight lost on Diet

for losing weight? diet (Scale) (Nominal)
Exercise: Solution
Research question Dependent Independent Test
variable variable
Do women do more Housework (hrs Gender Independent t-
housework than men? per week) (Nominal) test
(Scale)
Does Margarine X Cholesterol Occasion Repeated
reduce cholesterol? (Scale) (Nominal) measures ANOVA
Everyone has
cholesterol measured
on 3 occasions

Which of 3 diets is best Weight lost on Diet One-way ANOVA

for losing weight? diet (Scale) (Nominal)
Tests investigating relationships
Investigating Dependent Independent Test
relationships between variable variable

2 categorical variables Categorical Categorical Chi-squared test

2 Scale variables Scale Scale  Pearson’s correlation

Parametric or non-parametric?
Statistical tests fall into two types:
Assume data follows a
Parametric tests particular distribution e.g.
normal

Nonparametric techniques
Non-parametric are usually based on ranks/
signs rather than actual data
Parametric tests Non-parametric tests
 Statistical tests for  Distribution free statistics
population parameters
(means, variance,  Used in place of their
proportion) parametric counterparts
when the assumption of
 Basic assumption: normality can not be met
population is normally
distributed
Do these look normally distributed?
Do these look normally distributed?
yes

no yes
Non-parameteric tests
Non-Parametric Parametric Alternative
 The Sign Test  Z test

 Wilcoxon rank sum test or  Z test, t test (independent

Mann-Whitney U test samples)
 t test (dependent samples)
 Wilcoxon signed- rank test

 ANOVA
 Kruskal-Wallis test

 Spearman rank correlation

 Pearson coefficient
coefficient
Thanks