Validity in 1

Experimentation
GROUP 5 :
1. NAHIN RIFAT DIP ID : 202005
2. MD. LELINUR RAHMAN ID : 202012
3. RADHA SAHA ID : 202028
4. MD. ATIQUL ISLAM ID : 202035
5. MD. ABDUL QUAYYUM ID : 202038
 2

Presented By
NAHIN RIFAT DIP
ID – 202005
 Validity 3

 Validity refers to whether a study is able to

scientifically answer the questions it is intended to
answer.

 In short, Validity refers to how accurately a method

measures what it is intended to measure.
 Example of validity 4

 Drug Testing:
When testing the effectiveness of a new medication or
treatment, experimental validity ensures that the results
accurately reflect the medication's effects on the targeted
patient population. This requires careful control over factors
such as the study design, randomization, and blinding.

Besides this, we use validation in Educational Research,

Market Research, Environmental Research etc.
 Reliability 5

 Reliability refers to the extent to which a

measurement or experiment consistently produces
the same results under consistent conditions.

 In short, Reliability refers to the consistency, stability,

and repeatability of measurements, tests, or
research findings.
Validity Vs Reliability 6
 7

Presented By
MD. LELINUR RAHMAN
ID – 202012
 Types of Experimental Validity 8

 There are 2 major types of Experimental validity.

They are –
1. Internal Validity
2. External Validity
 Internal Validity 9
 Internal validity refers to the degree to which a research study
accurately establishes a cause-and-effect relationship between
dependent and independent variables.

 Example : It is lower in an inquiry that examines the link between

income level and the risk of smoking. According to a study,
there is a correlation between smoking and being a low-
income person. Occupation, culture, education, social
standing, and other variables are examples of different sorts of
factors. Such factors cannot be eliminated from the research.
 Internal Validity Threats 10

 8 kinds of confounding variable that can interfere with internal validity.

 These are -
1. History
2. Selection
3. Testing
4. Instrumentation
5. Maturation
6. Experimental mortality
7. Statistical Regression
8. Selection – maturation interaction
 History 11
 During the same time an experiment is taking place, some
unplanned event may occur that confuse the relationship being
studied.
 In other words we can say, Events that occur at the same time
which can change participants’ behavior.

Example : A researcher conducts a study to assess the impact of a

new teaching method on student performance in a mathematics
class. The researcher introduces the new teaching method to one
group of students while the other group receives traditional
instruction. At the end of the semester, the group using the new
method performs significantly better on the final exam.
 Selection 12

 Differences exist between individuals in experimental groups and

control groups at the start of the study.

 Examples : A researcher wants to assess whether a new teaching

method improves test scores in a chemistry class. The researcher
selects a group of students who are known to be highly motivated
and high-achieving for the experimental group, while a group of
struggling students is assigned to the control group.
The observed differences in test scores could be attributed to the
students' characteristics rather than the new teaching method.
 13

Presented By
RADHA SAHA
ID – 202028
 Testing 14

 Testing is the process of taking a pretest can affect the scores of a

postest.
 Example : A researcher is investigating how different study
techniques impact memory recall. Participants are divided into two
groups: one group studies a set of materials and takes a practice
quiz immediately afterward, while the other group only studies the
materials.
 The act of taking the practice quiz (testing) immediately after
studying could lead participants to remember the information
better on the final recall test, regardless of the effectiveness of the
study technique.
 Instrumentation 15

 Any change in the wording of questions, a change in interviewers

or a change in other procedures to measure the dependent
variable causes an instrumentation effect.

 Example : In a study investigating the effects of a new teaching

technique on students' comprehension, researchers initially assess
students' understanding of a topic through written essays. Later in
the study, they switch to assessing comprehension using multiple-
choice quizzes. The change from written essays to multiple-choice
quizzes introduces an instrumentation threat because the two
assessment methods may effect on results.
 Maturation 16

 The maturation effect is an effect on the results of an

experiment caused by changes in the experimental subject
over time.
 In short, Natural changes or developments in participants over
time that could affect the outcomes.

 Example : A study aims to investigate whether a new reading

program improves reading skills in elementary school students
over a school year. At the end of the year, the students' reading
scores have improved compared to the beginning.
 17

Presented By
MD. ATIQUL ISLAM
ID – 202035
 Experimental Mortality 18

 This occurs when the composition of the study groups changes

during the test.

 Example : A study is conducted to examine the effects of a new

exercise routine on weight loss over a two-month period.
Participants are randomly assigned to either an exercise group or a
control group. However, as the study progresses, several
participants drop out of the exercise group due to scheduling
conflicts and lack of motivation.
 Statistical Regression 19
 Statistical regression is a phenomenon where extreme or unusually
high/low scores on a measurement tend to move closer to the
average or mean is called statistical regression.

 Example: In a classroom, a teacher gives a surprise quiz and a

student who usually performs poorly scores much higher than usual.
On the next quiz, the student's score is not as high. This is an instance
of statistical regression. The initial extreme score is likely to regress
toward the student's typical level of performance over time, even
without any significant change in the student's knowledge or
abilities. It's important to recognize this effect when assessing the
effectiveness of teaching methods.
 Selection – maturation interaction 20
 A threat where the combination of non-random participant selection
and natural maturation processes affects the study outcomes. That’s
mean two factor influence the experimental results.
 Example : A study investigates the effects of a new study technique
on exam scores in two different student groups: honors students and
regular students. After implementing the study technique, the honors
students demonstrate a significant increase in exam scores, while the
improvement among regular students is less notable due to the
technique and their initial lower achievement level.
The selection of student groups (honors vs. regular) interacts with the natural
maturation of students' academic skills to create the observed differences in
outcomes.
 21

Presented By
MD. ABDUL QUAYYUM
ID – 202038
 External Validity 22

 External validity refers to the extent to which the results of a study or experiment
can be generalized beyond the specific conditions and participants of the study
to a broader population, different settings, and various times.

 In other words, it assesses the degree to which the findings from a study can be
applied or extended to real-world situations and different groups of people.

 Example : An education technology company conducts a study to assess the

effectiveness of an online learning platform in improving student performance. The
study is conducted with a group of college students taking a specific course
online. The company wants to know if the positive outcomes observed in this
course can be extended to students in other disciplines and academic levels.
 External Validity 23

 External Validity can be split into two distinct types. They are –
1. Population Validity
2. Ecological Validity.

1. Population Validity : This type of external validity assesses the extent to which the
results of a study can be generalized to a larger population beyond the sample
that was actually studied. It is also known as also known as generalizability.
Example :
Suppose a pharmaceutical company is conducting a clinical trial to test the
effectiveness of a new medication for treating a specific type of cancer in some
major cities and generalize to other city.
 External Validity 24

 Ecological Validity, also known as Environmental Validity, refers to the

extent to which the results an experiment can be generalized from a
set of environmental conditions created by the researcher to other
environmental conditions.
 it assesses whether the results obtained in a controlled laboratory or
experimental setting can be generalized and applied to natural or
everyday situations.
 Example : Imagine a study investigating the effects of noise pollution on
sleep quality. Researchers conduct their experiment in a soundproof
lab, where participants sleep while exposed to controlled levels of
noise.
Jeopardizing Factors 25

 Several factors can jeopardize external validity, making it difficult to

generalize research findings to broader populations, settings, and contexts.
They are-

1. Interaction effect of testing,

2. Interaction effect of selection bias and experimental variable,
3. Reactive effects of experimental arrangements,
4. Multiple treatment inference.
Jeopardizing Factors 26

 Interaction effect of testing : An interaction effect occurs when the

relationship between two variables changes based on the presence or level
of a third variable.
 Interaction effect of selection bias and experimental variable : The
interaction effect of selection bias refers to a situation in research where the
imbalance in the composition of a study's participants, leading to potentially
misleading or distorted conclusions about the relationship between variables.
 Interaction effects of experimental arrangements occur when the way an
experiment is designed and conducted (experimental arrangements)
influences the relationship between variables being studied.
 multiple treatment interference: In experimental designs requiring the
administration of more than one treatment to the same subject(s), the effect
of one treatment may be influenced by the effect of another treatment
 Internal VS external Validity 27
Internal External
1. The extent to which the study's results 1. The extent to which the study's findings
accurately reflect the relationship between can be generalized to different populations,
variables within the experiment. settings, and times.

2. Primarily concerned with the study's 2. Concerned with the broader context,
specific sample, variables, and experimental including different populations, settings, and
conditions. times.
3. Often takes place in a laboratory or 3. May occur in more naturalistic settings to
controlled environment to minimize increase external validity but with potential
extraneous influences. sacrifices in control.
4. Useful for establishing theoretical 4. Important for informing practical
foundations applications
Example: A controlled experiment testing the Example: A study assessing the effectiveness
effects of a new drug on a specific group of of a teaching method across different
participants in a laboratory setting. schools and regions to see if the findings hold
in various educational contexts.
 28

THANK YOU
ANY QUESTION ???