Scribd
Upload
2 views

Validity Reliability

The document discusses the concepts of reliability and validity in measurement, emphasizing their importance for accurate data collection and analysis. Reliability refers to the consistency of a measurement, while validity pertains to the accuracy of what is being measured. It outlines methods to enhance both reliability and validity, including appropriate measurement techniques, standardization, and careful sampling.

Uploaded by

Abdulrahman Mohammed
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
2 views

Validity Reliability

The document discusses the concepts of reliability and validity in measurement, emphasizing their importance for accurate data collection and analysis. Reliability refers to the consistency of a measurement, while validity pertains to the accuracy of what is being measured. It outlines methods to enhance both reliability and validity, including appropriate measurement techniques, standardization, and careful sampling.

Uploaded by

Abdulrahman Mohammed
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 13

RELIABILITY AND VALIDITY

Reliability and validity: Definition


2

 Concerned with measurement accuracy and error


 Thus enhance our ability find differences between populations and timepoints
 Subsequently provides results that are appropriate to make generalizations and
conclusions
 Increase transparency and minimize researcher bias
 Important for data collection, analysis, and interpretation.
 Reliability vs. validity: closely related but with different meanings.
 A reliable measurement can be valid or not valid.
 But valid measurement is usually reliable.
Enhancing validity and reliability
3

They are important thus the results reflect the real variations between
populations and timeframes as accurately as possible.
Shouldbe considered during the early stages and throughout data collection.
Choose appropriate methods of measurement
High quality and appropriate measurement technique to measure exactly what suppose
to be measured accurately.
If a new questionnaire is developed, it should be:
Based on previously established theory or study findings
Carefully and precisely worded.
 Use appropriate sampling methods to select the subjects
 Select sufficient, appropriate, and representative population and clearly
define it including age, location, gender and socioeconomic status.
Enhancing validity and reliability
4

 Give all participants the same information


 Precise, stable and reproducible tool/technique for collecting data is
important for generalizable results.
 Standardization: reduce the influence of external factors:
 Use the tools in a consistent manner including obtaining the measurements under
similar conditions
 Document the procedures, thus can be followed everytime the measurements are
obtained, especially when several testers are involved
 Procedures including measurement sequence and timing, environmental factors, diet, and
smoking.
 The reliability and validity should be discussed throughout the thesis sections,
but presented in details in the methods.
Reliability: Definition
5

 The consistency of a method/instrument/device/tool/scale to quantify a


measurement, when obtained under identical conditions/circumstances.
 A scale/device consistently obtaining the same value/score for a measurement
 Even if there is a margin of error (but it wouldn’t be valid)

 Exmp: the ability of automatic sphygmomanometers to repeatedly give the same


BP values when measured under identical conditions.

 Exmp: A diagnostic questionnaire giving different values under identical


conditions with every time used indicates low reliability.
Reliability: Definition
6

 Is estimated/assessed by comparing value/score varieties of the same scale.


 Expm: interclass correlations coefficient (ICCC)
 A very complex procedure:
 It describes the strength of resemblance between units in the same group.
 Uses correlation and comparison between measurements at several timepoints
Validity: Definition
7

Validity
 A device/scale accurately producing values correspond to the actual
properties, characteristics, and variations of the element being measured.
1. The accuracy of a method to quantify a measurement
2. The method must measure what it alleges/intends to measure
 Exmp: A thermometer should accurately measure temperatures. It is not valid if:
 It dosnt measure temp
 It gives inaccurate measurement.
 Exmp: A scale should accurately measure pain NOT QOL in LBP patients
Reliability vs. validity
8
Reliability vs. validity
9

 Validity is more important but harder to determine than reliability


 For useful results, discussion, and conclusions, a valid scale is a must
 Reliability on its own is not sufficient to secure validity.
 A reliable scale/device MAY NOT accurately reflect the real properties of the
element.
 Exmp 1:
 A thermometer can constantly give the same results (i.e. reliable).
 However, it may constantly give 2 degrees lower than the true value (i.e. may not be
properly calibrated), thus NOT valid.
 Exmp 2:
 A test designed to measure working memory correlates weakly with other memory tests
and strongly with reading comprehension, indicates low validity
 Thus this test may be appropriate for reading comprehension instead of memory.
Reliability: Types and assessments
10

Test-retest reliability
 The consistency of a scale across time: getting the same results when the
measurements repeated overtime
 The ability of a tool to get the same score of cognitive function repeatedly
overtime (e.g. after 1 day, 2 weeks, and 3 months).
Interrater reliability
 The consistency of a measure across testers: getting the same scores when
the measurement is conducted by different testers
 It is to assess the lab technician consistency
Reliability: Types and assessments
11

Internal consistency
 The consistency of the measurement itself: the correlation between scores
of different parts of a scale.
 Finding a correlation (ICCC) between the scores of randomly selected items
of a questionnaire designed to measure self-esteem
 It is to assess the device/scale consistency
Validity: Types and assessments
12

Construct
 Testing a scale/device against the “theory” upon which the scale was
constructed or based on
 i.e. testing the device against the theory
Content
 Ability of an instrument to reflect/represent all aspects/domains of the
instrument
 Exmp: a statistics teacher attempts to check the overall student knowledge in

the subject. A test with content validity:


 Should cover every topic of statistics that was covered in the semester.
 Shouldn’t contain unrelated/uncovered material (e.g. geography)
Validity: Types and assessments
13

Criterion
 Testing a newly-developed instrument against another valid instrument
 Exmp: comparing BP score using a smart watch versus the conventional method
 Exmp: ability of a survey to predict election results.
 To assess the validity of a cause-and-effect relationship, you also need to
consider internal validity (the design of the experiment) and external validity
(the generalizability of the results).
Face
 If the test “appears” (at face value) to measure what it claims to.
 The least sophisticated of all validity types.
 Exmp: asking people to rate the accuracy of a newly-developed weighing scale
as it appears to them using Likert scale (strongly agree, agree, disagree
strongly disagree)

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy