Professional Documents
Culture Documents
B337 L04 InteractiveSeminar Student
B337 L04 InteractiveSeminar Student
Interactive
Seminar
Learning Outcomes
Recap of Lesson 3
Research Design
Survey Method
Descriptive Statistics
Sampling Methods
OFFICIAL (CLOSED) \ NON-SENSITIVE
Discussion Revisited
Your school will be embarking on a “Happiness
Project” to enhance the mental wellbeing of
students by reducing the stress level of students.
Dr Nic's Maths and Stats. (2011). Hypothesis testing: step-by-step, p-value, t-test for difference of two means - Statistics Help.
Retrieved from https://www.youtube.com/watch?v=0zZYBALbZgg
OFFICIAL (CLOSED) \ NON-SENSITIVE
• Independent samples
– Values collected from two different groups
– Two types:
• Assuming equal variance / unequal variance
– Conduct an f-test to determine
• Paired samples
– Paired measurements taken from the same group of
people
• Samples are “related” to each other
OFFICIAL (CLOSED) \ NON-SENSITIVE
Comparing Samples
Consider this example:
Uncommon Uncommon
Outcomes Outcomes
Rare Rare
Outcomes Outcomes
mean
OFFICIAL (CLOSED) \ NON-SENSITIVE
Standard Deviation
If a distribution is normal, the following empirical rules apply:
• ~ 68% of all values in the distribution lie within 1 standard deviation
of the mean
• ~ 95% of all values in the distribution lie within 2 standard deviations
of the mean
• ~ 99.7%of all values in the distribution lie within 3 standard
deviations of the mean
Standard Deviation
• It roughly describes how far
away the typical observation is
from the mean. We call the
distance of an observation
from its mean its deviation.
Comparing Samples
Applying the normal distribution to the scenario:
RP
• The RP distribution will
look something like this:
109
OFFICIAL (CLOSED) \ NON-SENSITIVE
Comparing Samples
• The mean score of the XP sample in the IQ test is 109. The
diagram below shows RP’s IQ distribution, with the XP
sample’s mean IQ included.
• From this, we can see that at least 16% of all RP samples
performed above the XP mean.
RP XP
Based on the
sample, can you
conclude that XP
students are more
intelligent than RP
students?
100 109
OFFICIAL (CLOSED) \ NON-SENSITIVE
Comparing Samples
• Researchers typically use 1 – 10% as the limit
– Based on the distribution curve, what are the
chances of an RP student scoring higher or RP XP
equal to XP’s mean? Is it </= 10%?
– I.e., are the observed scores distinct
enough to assume that RP and XP
students are inherently different?
100 109
• In this case, conclude that XP students are not more intelligent than
RP students or intelligence levels between the students of RP and
XP are not significantly different.
– XP sample mean having a higher score than RP sample mean is
probably due to random error (luck of draw) into the sample.
OFFICIAL (CLOSED) \ NON-SENSITIVE
Comparing Samples
RP XP
• Now, assume that the XP sample’s
mean IQ is within the top 2.5%
of RP’s IQ distribution.
100 109
Discussion
We have learnt about descriptive statistics in the
previous lesson and inferential statistics today.
Discussion
The differences between Descriptive and Inferential statistics:
Source: https://www.cuemath.com/data/descriptive-and-inferential-statistics/
OFFICIAL (CLOSED) \ NON-SENSITIVE
Worksheet Question 1
Identify if each of the scenarios is a one-tailed or two-tailed
test and write possible hypotheses for each of them.
Statisticians love
whoever they spend the
most time with; that’s
their statistically
significant other.
Source: https://datasciencedojo.com/community/jokes/
OFFICIAL (CLOSED) \ NON-SENSITIVE
• Hypothesis:
• Compare the p value against the α value that was set (since this is a two-tailed test,
we look at “P(T<=t) two-tail)”
– In this case, we observe that p < α (0.028 < 0.05)
– The null hypothesis is therefore rejected when α = 0.05
Worksheet Question 3
Worksheet Question 3
Worksheet Question 3
Worksheet Question 3
T-Test ANOVA
The t-test is a statistical technique The ANOVA is a statistical technique
that is used to compare the means that is used to compare the means
of two groups of more than two groups
Worksheet Question 4
Zac recently read that the key to happy customers is happy employees –
employees who are satisfied are more likely to improve customer relationships,
resulting in increase in sales. He decided to administer the Positive and Negative
Affect Scale (PANAS) to the employees in the three teams of the Sales
Department to find out their level of positive affect. The three teams are the Inside
Sales Team, Outside Sales Team, and the Account Executive Team. The
composite scores are attached in the excel sheet below.
You have been assigned to find out whether there is any difference in the level of
positive effect across the three teams. This can be achieved by conducting a
One-Way Analysis of Variance (ANOVA), which is a statistical test meant to
compare values across groups.
In Grouped By,
choose columns.
Ensure that
everything is properly
filled out and click
“OK”.
OFFICIAL (CLOSED) \ NON-SENSITIVE
Worksheet Question 5
A monthly Happiness Survey was administered to a sample
of staff members monthly for the first five months this year. A
temp data entry staff has entered the composite scores into
an excel file attached below.
Ensure that
everything is
properly filled out
and click “OK”.
OFFICIAL (CLOSED) \ NON-SENSITIVE
Final Activity
Part 1)
Final Activity
Part 2)
Additional Resources
Dr Nic's Maths and Stats (4 Feb, 2019). Type 1 and Type 2 errors - Statistics Help. Retrieved from
https://www.youtube.com/watch?v=edzQQFNzFjM
OFFICIAL (CLOSED) \ NON-SENSITIVE
Conclusion
Concept Map
Inferential Statistics
T-test ANOVA
Independent
One-Way ANOVA
Samples
Repeated Samples
Paired Samples
ANOVA
OFFICIAL (CLOSED) \ NON-SENSITIVE
References