Skip to Main Content

Wellness Promotion -- M.A.

Helpful research instructions, tips, and links for the graduate, completely-online Wellness Promotion program.

Reliability vs. Validity

When evaluating research, you always need to consider "reliability" and "validity." 

Measurement tools can be:

  • Reliable, but not valid
  • Neither reliable nor valid
  • Both reliable and valid

You cannot have a measurement tool that is not reliable and not valid.  

IN ORDER TO BE VALID, A MEASUREMENT TOOL HAS TO BE RELIABLE. 

____________________________________________________________________________________________________________________

Reliability is:

  • the accuracy of the instrument
  •  the extent to which a research instrument consistently has the same results if it is used in the same situation on repeated occasions.
  • the measurement could apply to others/the same result would happen each time
  • Dependability or consistency

Reliability in quantitative research:

Test-retest The same test over time.
Interrater The same test conducted by different people.
Parallel forms Different versions of a test which are designed to be equivalent.
Internal consistency                       The individual items of a test.

 

Reliability in qualitative research:

Credible Has internal validity. 
Transferable Is it important to consider the findings of the study within the broader context of other people and settings, and whether similar projects and methods conducted in different environments would be of value.
Dependable  The processes within the study are reported in detail, so that a future researcher can repeat the work
Findings can be confirmed                       Steps are taken to make sure the work’s findings are the result of experiences and ideas of the informants, and that evaluation findings are arrived at by considering solid evidence.

 

Middleton, F. (2021 July 16). Types of reliability and how to measure them. Scribbr. https://www.scribbr.com/methodology/types-of-reliability/#:~:text=Internal%20consistency%20assesses%20the%20correlation,only%20have%20one%20data%20set.

NSF Consulting. (2021). The reliability of qualitative research. NSF Consulting Pty Ltd. http://nsfconsulting.com.au/qualitative-research-reliability/

____________________________________________________________________________________________________________________

 

Validity is:

  • a term describing a measure that accurately reflects the concept it is intended to measure
  • truthfulness

In quantitative studies:

  • Face validity:  the study seems to be a reasonable measure of some variable
  • Content validity:  the degree to which a measure covers the range of meanings included with a concept
  • external validity:  the degree to which findings are generalizable to a population
  • criterion-related validity

In qualitative studies:

  • Questions are clear
  • The researchers role and status are described
  • the findings show meaningful parallelism across data sources
  • If multiple field workers, they have comparable data collection protocols.  
  • coding checks made, and they showed adequate agreement

       You can increase validity in qualitative research by:

  • prolonging the engagement
  • purposeful sampling
  • triangulation = using multiple data source
  • member checks

F. Oguz, personal communication, May 29, 2019.