CHAPTER 1 - LESSON 1
Reliability vs. Validity
Learning Objectives
Students should be able
to identify the concept of
reliability and validity in
the context of research
and measurement.
Students should be able
to differentiate between
various types of
reliability, such as test-
retest reliability, inter-
rater reliability, parallel
forms reliability, and
internal consistency.
Students should be
able to apply the
concepts of reliability
and validity to different
fields, such as
psychology, education,
medicine, and
business.
1 2 3
RELIABILITY
VALIDITY
&
If you use the same
research instrument
under the same
conditions, do you get
consistent results?
Reliability
It refers to the repeatability or consistency of
your measurement.
WHAT IS RELIABILITY?
Reliability
It also refers to whether research methods can
reproduce the same results multiple times. If
your research methods can produce consistent
results, then the methods are likely reliable and
not influenced by external factors.
EXAMPLES: Classroom Test
Reliability
Reliable Test
• If a student takes the test today and scores 85%,
and then takes the same test next week under
similar conditions and also scores around 85%,
the test is reliable. The scores are consistent over
time.
Unreliable Test
• If the same student scores 85% today but scores
50% next week without any change in their
knowledge or preparation, the test might be
unreliable. The inconsistency in scores suggests
that the test may not accurately reflect the
student's understanding.
How can we measure reliability?
Over Time:
Test-Retest Reliability
Let’s have a think about the different ways in which we might measure reliability:
Across different researchers :
Inter-Rater Reliability
Across different tool:
Parallel Forms Reliability
Across different items:
Internal Consistency Reliability
Do you get the
same results
when you repeat
the
measurement?
When doing an experimental,
Test-Retest
Reliability
(Over Time)
It involves assessing the consistency of
measurements over time. It’s like taking
the same measurement or test twice –
once and then again after a certain period.
If the results align closely, it indicates that
the measurement is reliable over time.
Think of it as capturing the essence of
stability.
WHAT IS TEST-RETEST RELIABILITY?
Test-Retest
Reliability
(Over Time)
Do you get the same
results when
different people
conduct the same
measurement?
When you conduct your experimental study,
Inter-Rater
Reliability
(Across different
researchers)
This type of reliability assesses the level of
agreement between different observers
when evaluating the same phenomenon.
This relates to consistency across people. If
someone else repeats the measurement
(e.g., someone else rates my intelligence)
will they produce the same answer?
WHAT IS INTER-RATER RELIABILITY?
Inter-Rater
Reliability
(Across different
researchers)
It measures the correlation between two
equivalent versions of a test. You use it when you
have two different assessment tools or sets of
questions designed to measure the same thing.
WHAT IS PARALLEL FORMS
RELIABILITY?
Parallel Forms
Reliability
(Across different tool)
Example:
If the same students take two different versions
of a reading comprehension test, they should
get similar results in both tests.
When you conduct your study,
Internal Consistency
Reliability
(Across Items)
Do you get the same
results from different
parts of a test that are
designed to measure the
same thing?
It assesses the correlation between
multiple items in a test that are intended to
measure the same construct or topic.
WHAT IS INTERNAL CONSISTENCY
RELIABILITY?
Internal Consistency
Reliability
(Across items)
Example:
If you are designing a test on geometry,
then all questions on the test should be
about geometry.
Cronbach’s Alpha
when designing and testing a
new survey or assessment
instrument
helps them evaluate the quality of
the tool during the design phase
before deploying it fully.
before including various questions
in a scale, they must be sure that all
items reliably measure the same
construct
to ensure that items agree, but they
need to use it with other analyses that
evaluate whether the items measure
the correct characteristic.
Does the research
instrument measure
what it’s supposed to
measure?
Validity
Validity is an evaluation of how accurate the
study is. It describes the extent to which the
study actually measures what it intends to
measure.
WHAT IS VALIDITY?
Validity
Example:
A valid driving test should include a practical
driving component and not just a theoretical test
of the rules of driving.
Some Example of Validity Testing
Content Validity: Used when you want to know whether a sample of items truly
reflects an entire universe of possible items in a certain topic.
Example: “I want to design a test to measure my students’ ability in statistics”
Q1: Calculate the mean of the sample.
Q2: Calculate the mean of the other sample.
Q3: Calculate the mean of the other sample.
Q4: Calculate the mean of the other sample.
Q5: Calculate the mean of the other sample.
Q1: Calculate the mean.
Q2: Calculate the standard deviation.
Q3: Calculate the correlation.
Q4: Calculate a t-test.
Q5: Calculate an ANOVA.
Some Example of Validity Testing
Criterion Validity: Used when you want to assess
whether a test reflects a set of abilities in a
current or future setting.
a.Concurrent Validity: Checks how well a test
correlates with an existing measure at the same
time.
b.Predictive Validity: Checks how well a test predicts
future performance or behavior.
Some Example of Validity Testing
Construct Validity: Does the
instrument measure the
correct attribute?
Reliability vs. Validity
Research
Instruments
• Research-Made
• Standardized
• Modified Standardized
Research-Made Instrument:
How research-made instruments
related to Validation and Reliability
Research-Made
A researcher-made instrument refers to a
measurement tool that is created by the researcher
specifically for their study. These instruments are
tailored to the unique characteristics and objectives
of the research. They are designed to gather data
relevant to the specific research questions, variables,
or phenomena under investigation.
Research-Made Instrument:
How research-made instruments
related to Validation and Reliability
Research-Made
Research-made instruments offer flexibility, allowing
researchers to customize questions and response
formats based on the study’s focus. However, they
need careful development to ensure validity and
reliability.
VALIDATION: YES RELIABILITY: YES
Research-Made Instrument:
How standardized instruments related
to Validation and Reliability
STANDARDIZED
INSTRUMENT The standardized instruments is a pre-existing
measurement tool that has been developed and
validated by researchers, often through rigorous
testing and standardized procedures. These
instruments have established reliability and validity
and are designed to be administered consistency
across different settings and populations
Research-Made Instrument:
How standardized instruments related
to Validation and Reliability
STANDARDIZED
INSTRUMENT Standardized instruments enable researchers to
make comparisons across studies and populations,
enhancing the generalizability of findings. Examples
include established psychological assessments,
educational Test, or health surveys.
VALIDATION: NO RELIABILITY: NO
Research-Made Instrument:
How modified-standardized instruments
related to Validation and Reliability
MODIFIED-
STANDARDIZED
INSTRUMENT A modified standardized instruments is a pre-
existing measurement tool that has been altered or
adapted better suit the specific context or
population of a research study. Researchers may
make modifications to ensure cultural relevance,
linguistics appropriateness, or to address specific
research objectives.
Research-Made Instrument:
How modified-standardized instruments
related to Validation and Reliability
MODIFIED-
STANDARDIZED
INSTRUMENT While modification can enhance the instrument’s
applicability to the study, researchers must be
cautious to maintain the validity and reliability of the
original instruments after any alterations.
VALIDATION: YES RELIABILITY: YES
Remember
A valid instrument is a reliable
instrument, while a reliable
instrument is not always a
valid instrument.
Thank you for
listening!

LESSON 1 RELIABILITY AND VALIDITY for Statech12

  • 1.
    CHAPTER 1 -LESSON 1 Reliability vs. Validity
  • 2.
    Learning Objectives Students shouldbe able to identify the concept of reliability and validity in the context of research and measurement. Students should be able to differentiate between various types of reliability, such as test- retest reliability, inter- rater reliability, parallel forms reliability, and internal consistency. Students should be able to apply the concepts of reliability and validity to different fields, such as psychology, education, medicine, and business. 1 2 3
  • 3.
  • 4.
    If you usethe same research instrument under the same conditions, do you get consistent results? Reliability
  • 5.
    It refers tothe repeatability or consistency of your measurement. WHAT IS RELIABILITY? Reliability It also refers to whether research methods can reproduce the same results multiple times. If your research methods can produce consistent results, then the methods are likely reliable and not influenced by external factors.
  • 6.
    EXAMPLES: Classroom Test Reliability ReliableTest • If a student takes the test today and scores 85%, and then takes the same test next week under similar conditions and also scores around 85%, the test is reliable. The scores are consistent over time. Unreliable Test • If the same student scores 85% today but scores 50% next week without any change in their knowledge or preparation, the test might be unreliable. The inconsistency in scores suggests that the test may not accurately reflect the student's understanding.
  • 7.
    How can wemeasure reliability? Over Time: Test-Retest Reliability Let’s have a think about the different ways in which we might measure reliability: Across different researchers : Inter-Rater Reliability Across different tool: Parallel Forms Reliability Across different items: Internal Consistency Reliability
  • 8.
    Do you getthe same results when you repeat the measurement? When doing an experimental, Test-Retest Reliability (Over Time)
  • 9.
    It involves assessingthe consistency of measurements over time. It’s like taking the same measurement or test twice – once and then again after a certain period. If the results align closely, it indicates that the measurement is reliable over time. Think of it as capturing the essence of stability. WHAT IS TEST-RETEST RELIABILITY? Test-Retest Reliability (Over Time)
  • 10.
    Do you getthe same results when different people conduct the same measurement? When you conduct your experimental study, Inter-Rater Reliability (Across different researchers)
  • 11.
    This type ofreliability assesses the level of agreement between different observers when evaluating the same phenomenon. This relates to consistency across people. If someone else repeats the measurement (e.g., someone else rates my intelligence) will they produce the same answer? WHAT IS INTER-RATER RELIABILITY? Inter-Rater Reliability (Across different researchers)
  • 12.
    It measures thecorrelation between two equivalent versions of a test. You use it when you have two different assessment tools or sets of questions designed to measure the same thing. WHAT IS PARALLEL FORMS RELIABILITY? Parallel Forms Reliability (Across different tool) Example: If the same students take two different versions of a reading comprehension test, they should get similar results in both tests.
  • 13.
    When you conductyour study, Internal Consistency Reliability (Across Items) Do you get the same results from different parts of a test that are designed to measure the same thing?
  • 14.
    It assesses thecorrelation between multiple items in a test that are intended to measure the same construct or topic. WHAT IS INTERNAL CONSISTENCY RELIABILITY? Internal Consistency Reliability (Across items) Example: If you are designing a test on geometry, then all questions on the test should be about geometry.
  • 15.
    Cronbach’s Alpha when designingand testing a new survey or assessment instrument helps them evaluate the quality of the tool during the design phase before deploying it fully. before including various questions in a scale, they must be sure that all items reliably measure the same construct to ensure that items agree, but they need to use it with other analyses that evaluate whether the items measure the correct characteristic.
  • 16.
    Does the research instrumentmeasure what it’s supposed to measure? Validity
  • 17.
    Validity is anevaluation of how accurate the study is. It describes the extent to which the study actually measures what it intends to measure. WHAT IS VALIDITY? Validity Example: A valid driving test should include a practical driving component and not just a theoretical test of the rules of driving.
  • 18.
    Some Example ofValidity Testing Content Validity: Used when you want to know whether a sample of items truly reflects an entire universe of possible items in a certain topic. Example: “I want to design a test to measure my students’ ability in statistics” Q1: Calculate the mean of the sample. Q2: Calculate the mean of the other sample. Q3: Calculate the mean of the other sample. Q4: Calculate the mean of the other sample. Q5: Calculate the mean of the other sample. Q1: Calculate the mean. Q2: Calculate the standard deviation. Q3: Calculate the correlation. Q4: Calculate a t-test. Q5: Calculate an ANOVA.
  • 19.
    Some Example ofValidity Testing Criterion Validity: Used when you want to assess whether a test reflects a set of abilities in a current or future setting. a.Concurrent Validity: Checks how well a test correlates with an existing measure at the same time. b.Predictive Validity: Checks how well a test predicts future performance or behavior.
  • 20.
    Some Example ofValidity Testing Construct Validity: Does the instrument measure the correct attribute?
  • 21.
  • 22.
  • 23.
    Research-Made Instrument: How research-madeinstruments related to Validation and Reliability Research-Made A researcher-made instrument refers to a measurement tool that is created by the researcher specifically for their study. These instruments are tailored to the unique characteristics and objectives of the research. They are designed to gather data relevant to the specific research questions, variables, or phenomena under investigation.
  • 24.
    Research-Made Instrument: How research-madeinstruments related to Validation and Reliability Research-Made Research-made instruments offer flexibility, allowing researchers to customize questions and response formats based on the study’s focus. However, they need careful development to ensure validity and reliability. VALIDATION: YES RELIABILITY: YES
  • 25.
    Research-Made Instrument: How standardizedinstruments related to Validation and Reliability STANDARDIZED INSTRUMENT The standardized instruments is a pre-existing measurement tool that has been developed and validated by researchers, often through rigorous testing and standardized procedures. These instruments have established reliability and validity and are designed to be administered consistency across different settings and populations
  • 26.
    Research-Made Instrument: How standardizedinstruments related to Validation and Reliability STANDARDIZED INSTRUMENT Standardized instruments enable researchers to make comparisons across studies and populations, enhancing the generalizability of findings. Examples include established psychological assessments, educational Test, or health surveys. VALIDATION: NO RELIABILITY: NO
  • 27.
    Research-Made Instrument: How modified-standardizedinstruments related to Validation and Reliability MODIFIED- STANDARDIZED INSTRUMENT A modified standardized instruments is a pre- existing measurement tool that has been altered or adapted better suit the specific context or population of a research study. Researchers may make modifications to ensure cultural relevance, linguistics appropriateness, or to address specific research objectives.
  • 28.
    Research-Made Instrument: How modified-standardizedinstruments related to Validation and Reliability MODIFIED- STANDARDIZED INSTRUMENT While modification can enhance the instrument’s applicability to the study, researchers must be cautious to maintain the validity and reliability of the original instruments after any alterations. VALIDATION: YES RELIABILITY: YES
  • 29.
    Remember A valid instrumentis a reliable instrument, while a reliable instrument is not always a valid instrument.
  • 30.

Editor's Notes

  • #5 Reliability in research refers to the consistency and stability of the results obtained from a study. If a research method is reliable, it means that if the same study is repeated under the same conditions, it should produce similar results each time. Reliability is essential because it ensures that the findings of a study are trustworthy and can be replicated by other researchers.
  • #8 -consistency of a measure across time -Do you get the same results when you repeat the measurement? -Example: Imagine you create a math quiz for your students. You give them the quiz today and then again next week. If the students’ scores are similar both times, the quiz has high test-retest reliability.
  • #9 -consistency of a measure across time -Do you get the same results when you repeat the measurement? -Example: Imagine you create a math quiz for your students. You give them the quiz today and then again next week. If the students’ scores are similar both times, the quiz has high test-retest reliability.
  • #10 -consistency of a measure across raters or observers -Do you get the same results when different people conduct the same measurement? -Suppose two teachers are grading the same set of student essays. If both teachers give similar grades for each essay, there is high inter-rater reliability. However, if one teacher consistently gives higher grades than the other, the inter-rater reliability is low.
  • #11 -consistency of a measure across raters or observers -Do you get the same results when different people conduct the same measurement? -Suppose two teachers are grading the same set of student essays. If both teachers give similar grades for each essay, there is high inter-rater reliability. However, if one teacher consistently gives higher grades than the other, the inter-rater reliability is low.
  • #12 -Example: You create two different versions of a history test with similar types of questions but different wording. You give one version to the students in the morning and the other version in the afternoon. If the scores are similar, the test has high parallel-forms reliability.
  • #13 -consistency of the measurement -which all items or questions on a test or survey that are intended to measure the same concept yield similar results -Internal consistency reliability is important because it ensures that all items in a test or survey are measuring the same concept, which makes the results more valid. If a survey has high internal consistency, you can be more confident that the scores reflect what you are intending to measure.
  • #14 -consistency of the measurement -which all items or questions on a test or survey that are intended to measure the same concept yield similar results -Internal consistency reliability is important because it ensures that all items in a test or survey are measuring the same concept, which makes the results more valid. If a survey has high internal consistency, you can be more confident that the scores reflect what you are intending to measure.
  • #15 Spearman brown formula
  • #22 Research-made, Standardized, and Modified Standardized research instruments in Relation to VALIDITION and RELIBIALITY