Understanding Measurement Validity

you imagine that you wake up one morning with a cold you're coughing and sneezing and just generally feeling icky you think you might have a fever but you're not sure you get out your thermometer and stick it in your mouth but when you take it out the thermometer reads 72 degrees which is the temperature in your house not the temperature of your body in psychology there are many different tools used to measure trades surveys observations brain scans saliva samples these are just a few of the many different tools used in psychological measurement but what happens when a tool doesn't measure what it's supposed to measure validity is the extent to which a measurement tool measures what it's supposed to measure remember your thermometer it's measuring the room temperature not your body temperature since it's supposed to be measuring your body temperature the thermometer is not valid in the case of the thermometer we can easily see why validity is important without it we don't know what our body temperature is but what about in the case of psychological measurement for example imagine that we want to measure how empathetic a person is we write a survey to find out how much people understand and feel emotions for the plight of others if our survey has a high validity it does a good job of measuring empathy and people if it has a medium level of validity it might be measuring something else like altruism or guilt motivated behavior while both of these might be linked to empathy they're not empathy and if our survey has low validity it might be measuring something else entirely like intelligence or extraversion but even if it's measuring extraversion or intelligence instead of empathy who cares why is the Lydda T important if we think that we are measuring one thing but we're actually measuring something else then it won't give us the information we need just as your thermometer didn't tell you whether or not you have a fever our survey won't tell us how much empathy people have which is what we want to know but why wouldn't every cycle logical measure have validity after all we are writing the survey why don't we just write it to measure empathy that's easier said than done because empathy is a construct or idea that cannot be directly observed think about it like this you can't see or touch or taste empathy you can observe behaviors that indicate empathy but not imp a--they itself for example maybe you see someone giving a dollar to a homeless man or helping an old woman cross the street these behaviors hints that the person might have empathy but empathy isn't like the color of someone's hair you can't look at it and say oh yes I can see that so-and-so has blonde empathy today almost all psychological traits are constructs intelligence depression empathy prejudice resilience these are all psychological traits and none of them can be directly observed instead psychologists have to look at the people's words and behaviors and draw inferences about how those words and behaviors might relate to the underlying construct but what if what you think is related to an underlying construct really isn't for example what if the person we saw giving the homeless guy a dollar is helping because he wants to impress his date not because he feels empathy what if the person helping the old woman across the street is only doing it to clear the crosswalk so she can drive through it because we are always making inferences and those inferences can be wrong some measurement tools end up with low validity there are several major types of validity number one face validity is when a tool subjectively appears to measure a construct for example if we give a survey to people to test how empathetic they are we might look at the survey and say yeah this looks like it measures empathy face validity is not a good way to know if a measure is actually valid or not it involves people's opinions and opinions can be wrong number two construct validity is the extent to which a tool measures a construct for example if our survey actually measured empathy it would have a high construct validity because it is actually measuring the construct of empathy however construct validity is hard to prove as we've seen many things can look like it bethey so construct validity isn't always easy to demonstrate number three content validity is the extent to which the individual items on a test are relevant to the content area that it is testing if our survey had a question about whether a person likes spaghetti or not it wouldn't have a high content validity because liking spaghetti has nothing to do with empathy likewise you wouldn't put a spelling question on a math test that just isn't relevant notice that a tool can have high content validity and low construct validity our survey might ask questions that are all relevant to empathy and therefore have a high content validity but if it is measuring something other than empathy like guilt motivated behavior its construct validity is low number for predictive validity is when a tool can predict certain behaviors for example people who score high on our empathy survey might be more likely to volunteer regularly the people who score low on the survey we can predict what they will do based on their responses to the survey validity is the extent to which a measurement tool measures what it is supposed to measure psychologists need high validity on the assessments they use so that they know they are studying what they are trying to study however it is difficult to have high validity because most psychological traits are constructs or abstract ideas that are not observable there are four main types of validity face validity is the extent to which a tool appears to measure what it is supposed to measure construct validity is the extent to which a tool measures an underlying construct content validity is the extent to which items are relevant to the content being measured predictive validity is the extent to which responses on a measure can predict future behavior

Channel: Prof. Nikki Hozack

Views: 101,613

Rating: 4.9411764 out of 5

Keywords:

Id: kkjjZtFV9ZE

Length: 6min 50sec (410 seconds)

Published: Sat Mar 07 2015