Transcript Enhancing Thinking Skills in Science Context Enhancing Lesson 5

Enhancing Thinking Skills in
Science Context
Lesson 5
Double--blind Design in
Randomised blind Experimental Design
Activity 1
To study the effect of
coffee on humans’
concentration power
Hypothesis:
Coffee may enhance the
concentration power of
humans.
Materials and apparatus
• 1 packet of coffee A
• 1 packet of coffee B
• Enough hot water (at 800C ) for about 45
paper cups (half full)
• 1 bottle of coffee creamer
• 50 spoons
• 1 bag of red rice (net weight: 1.5 kg)
• 1 bag of white rice (net weight: 2 kg)
• 45 Petri dishes
• 45 Heat resistant paper cups
Procedure:
a) The class is randomly divided into three groups
(A, B and C), each of which consisting of roughly
the same number of students.
b) Each of you will be given a Petri dish containing
well--mixed red and mixed rice in the ratio 2:1
(approx.).
c) Each of you is required to pick out the red rice
from the white rice by hand at constant speed.
d) Each of you is required to record down the
number of red rice picked up in five minutes.
e) Record down the total no. of red rice picked up by
the different groups of students in five minutes.
f) Then each group will be given drinks according to
their group number immediately after step (e):
Group A: coffee A + coffee creamer [Drink A]
(3 spoonful of coffee A + 2 spoonful of coffee creamer
+half cup of hot water (800C) + stirring)
Group B: coffee B + coffee creamer [Drink B]
(3 spoonful of coffee B + 2 spoonful of coffee creamer
+ half cup of hot water (800C) + stirring)
Group C: water + coffee creamer [Drink C]
(2 spoonful of coffee creamer
+ half cup of hot water (800C) + stirring)
Procedure:
(g) Thirty minutes after taking the
drink each of you will be each of you
will be required to repeat steps (b),
(c), (d) and (e).
Result
Group A (Taken Drink A)
Student
1
2
3
4
5
6
First counting
experiment before taking
the drink:
No. of red rice picked up
from a mixture of red
and white rice in 5
minute
Second counting
experiment after taking
the drink:
No. of red rice picked up
from a mixture of red
and white rice in 5
minutes
Result
Group B (Taken Drink B)
Student
1
2
3
4
5
6
First counting
experiment before
taking the drink:
No. of red rice picked
up from a mixture of
red and white rice in 5
minute
Second counting
experiment after taking
the drink:
No. of red rice picked
up from a mixture of
red and white rice in 5
minutes
Result
Group C (Taken Drink C)
Student
1
2
3
4
5
6
First counting
experiment before
taking the drink:
No. of red rice picked
up from a mixture of
red and white rice in 5
minute
Second counting
experiment after taking
the drink:
No. of red rice picked
up from a mixture of
red and white rice in 5
minutes
Task:
1. Why was the experiment repeated?
(i.e. why do we need to do the first
counting experiment and then do the
second experiment? )
Task:
2. Why was the second experiment
carried out 30 minutes after taking
the drink?
Task:
3. Pool the group and class data after
the experiment. You may need to do
some calculation in order to obtain
more conclusive result. Show the
details of your treatment of the
class data.
Task:
4. Compare the results of group A, B
and C obtained from (3).Does the
result obtained support the
hypothesis that coffee may enhance
the concentration power of humans?
Please explain your answers.
The following data of the above experiment was
obtained from 3 classes (105 Secondary 2
students) of a school.
Compare the results and try to draw a conclusion.
Class data obtained from
2A, 2B, 2C
Discussion
of Experimental Results
a) Standard deviation indicates how spread out a set
of data is, that is, whether the data are relatively
close together and clustered around the mean or
spread out.
b) Referring to the standard deviation difference of
Class 2A, 2B and 2C the difference in standard
deviation of Class 2B is the greatest for Drink A and
Drink B, which makes the data collected in class 2B
not so trustworthy. However, the differences in
standard deviation of class 2A and 2C are
reasonable. Hence, the result of Class 2B was
discarded but the results of Class 2A and 2C are
combined to see whether there is any significant
evidence for drawing a conclusion.
Average of Percentage Increase of
Mean (Class 2A and 2C)
Control
Drink A
Drink B
16.6%
22%
39.3%
5.4
22.7
If set 0,
Average of Percentage Increase of
Mean (Class 2A and 2C)
Control
Drink A
Drink B
16.6%
22%
39.3%
If set 0,
5.4
22.7
(slightly greater
than that of the
Control)
(much greater
than that of the
Control and of the
“Drink A”group)
Drink B is decaffeinated coffee
• It contains nearly no caffeine
(only 0.1% caffeine).
• It was supposed to be used as a placebo.
A Double--blind Study
• A double-blind study, relatively easy to achieve in
drug studies, belongs to the category of Randomised
Experimental Design. Two similar groups of people
(experimental and control) act as targets of the
study. The experimental group is given an actual
treatment (e.g. an investigational drug) while the
control group is given a placebo (may be a sugar
pill), which has identical appearance (e.g. colour,
taste, size) as the investigational drug.
• A placebo is a preparation which is
pharmacologically inert but which may have a
medical effect solely on the power of suggestion, a
response known as the placebo effect.
A Double--blind Study
• Furthermore, people are randomly assigned to
the control or experimental group and given
random numbers by a study co-ordinator, who
also encodes the investigative drug and the
placebo with matching random numbers. Neither
the experimental subjects (both the control and
experimental group) nor the researchers
monitoring the outcome of the study know who
is receiving which treatment (the investigative
drug or the placebo), until the study is over and
the random code is broken. Such an
arrangement would prevent the subjects from
biasing the result by knowing which group they
are in.
A Double--blind Study
• If later it is found that results from the
experimental group is better than that
of the control group, the treatment or
the drug given to the experimental
group is assumed to have an effect.
• Finally, the same experiment (study)
need to be replicated on a large group
of subjects in order that more
significant conclusion can be drawn
about the effect of the investigative
drug.
Average of Percentage Increase of Mean (Class 2A and 2C)
Control
Drink A
Drink B
16.6%
22%
39.3%
If set 0,
5.4
(slightly greater than that
of the Control)
22.7
(much greater than that
of the Control and of
the “Drink A”group)
Drank coffee during the
experiment
Drank decaffeinated
(only 0.1% caffeine)
coffee during the
experiment
Implication: The result obtained from the Control, Drink A Group and Drink
B Group is not conclusive ((a) The result of Drink A group is only slightly
greater than that of the Control (not a significant difference); (b) The
placebo effect of Drink B Group should not be so great when compared with
Drink A Group and the Control).
Discussion Questions
1) Is the experimental design a double-blind
design? Explain your answer.
2) Is the experimental design a valid design?
Explain your answer.
3) If the experimental design is not a valid
one, suggest ways to improve the
experimental design
Suggested answers:
1) Is the experimental design a doubleblind design? Explain your answer.
The experimental design is a singleblind , not a double-blind because the
teacher knows the details of the drink
and which group of students would
take which drink.
2) Is the experimental design a valid design? Explain your answer.
Judging from the Percentage Increase (Mean) of Drink A,
B and the Control of Class 2A and 2C the experimental
design is not a valid design. Possible reasons may be as
follows:
• The sample size is too small;
• The suggested method is not a valid method to measure
the concentration power of humans;
• Some students may drink coffee during lunch or
breakfast and hence may interfere with the experimental
result
• Some students are more skillful at doing such task
requiring good motor skills;
• The effect of coffee is not significant because the “dose”
is not enough.
3) If the experimental design is not a valid one, suggest
ways to improve the experimental design?
Ways to improve the experimental design may be as follows:
•
•
•
•
•
•
To increase the sample size (invite more classes to do the
experiment);
To use another method to measure the concentration power of
the experimental subjects, e.g. Use photographic memory method
(eidetic imagery, in which a person memorises details of a
complex scene at a glance) to measure the concentration power
of the experimental subjects
Scattered numerals would appear for up to two-thirds of a second
in the computer screen and were then masked by white squares.
Each student is required to write down the numerals on a
worksheet according to their previous location within a fifth of a
second.
The same procedure with 10 different sets of numerals repeats
for 10 times and the students would check the answers after the
activity.
The percent accuracy of each student is counted and then the sum,
the mean, percentage increase, standard deviation are collected
for comparison among the three groups of students drinking the
control, drink A and drink B. as before.