Topic 3: Validity {by 6/4}

Based on the text readings and lecture recording due this week consider the following two discussion points: (1) Discuss your understanding of criterion-related validity (also known as: Prediction or Instrument-Criterion).  In you discussion, include why this particular type of validity is common/important for mental health assessments.  (2) First, discuss the difference between convergent evidence validity and discriminant evidence validity.  Second, provide an example of a hypothetical or real assessment where these two types of validity would be important – very common for many mental health assessments (hint: listen to my example in the lecture recording).

 

Your original post should be posted by 6/4.  Post your two replies no later than 6/6.  *Please remember to click the “reply” button when posting a reply.  This makes it easier for the reader to follow the blog postings.

33 Comments (+add yours?)

  1. Madi
    Jun 01, 2020 @ 10:29:23

    1. Criterion-related validity, also known as prediction or instrument-criterion, predicts future behavior. This kind of instrument is focused on the outcome, giving it a predictive component. Criterion-related validity is very important for mental health assessment because it allows mental health counselor to be able to predict future behaviors which allows them to decide which treatment to apply to which client. In being able to predict future behavior a counselor can also better intervene and help the client.

    2. Convergent evidence validity is when an instrument is related to other variables when it should be. This means that two instruments should have high validity. An example of this is that a new depression instrument should have high validity with an established depression instrument. Discriminant evidence validity is when an instrument is not correlated with variables from which it should differ from. This means that the two instruments should have low validity. An example of this is a depression instrument and a GAD instrument should have low validity since they are not measuring the same thing.

    Reply

    • Althea Hermitt McPherson
      Jun 04, 2020 @ 02:30:22

      Hi Madi, I like that you went a step further and spoke about using prediction to aid treatment goals and future intervention. I think that criterion validity is common because we all want to be able to predict when, where, how, why as these are common questions we seek to find answers to in our everyday life. I also liked your examples for convergent evidence and discriminant evidence validity, your explanation of both was simple, easy to understand and to the point.

      Reply

    • Yen Pham
      Jun 05, 2020 @ 13:42:40

      Hi Madi,
      I like your explanation on the question one. I consent with you that criterion-related validity predicts future behavior and it is focused on the outcome. However, I think it also is an index of how well a test correlates with an established standard of comparison (i.e., a criterion). Criterion validity is divided into two types: predictive validity, and concurrent validity. For example, if a measure of criminal behavior is valid, then it should be possible to use it to predict whether an individual (a) will be arrested in the future for a criminal violation, (b) is currently breaking the laws.

      Reply

    • Trey Powers
      Jun 07, 2020 @ 15:09:14

      Hi Madi!

      I found your description of the difference between convergent and discriminant validity to be quite helpful. Your description of when you would want high verses low validity in each case was clear, and I appreciated the fact that you provided examples for each type of validity.

      Reply

  2. Michelle McClure
    Jun 03, 2020 @ 16:43:24

    Criterion-related validity is how well an instrument is related to an outcome criterion, or how well it predicts future behavior or outcomes. This is important for mental health assessments, such as your example of suicide predictability, that is an important domain that the mental health field deals with regularly unfortunately. If you use a depression or suicidal scale as a mental health assessment then how well it predicts current and future behavior of the client is very important for the safety and mental health of the client.
    The difference between convergent evidence validity and discriminant evidence validity is that they are complete opposites and you want opposite validity coefficients for each one. Convergent evidence validity is when an instrument is related to other variables to which it should be positively related this would have a high validity coefficient and this is important when you are developing an instrument and you correlate it with an established instrument you want a high validity coefficient. Discriminant evidence validity is when an instrument is not correlated with variables from which it should differ, in this case you want a low validity coefficient so that it discriminates between variables. An example of when discriminant evidence would be important when you want to make sure your instrument is measuring what you want it to measure and not something else, for example you would not want a depression instrument to measure anxiety.

    Reply

    • Althea Hermitt McPherson
      Jun 04, 2020 @ 02:46:08

      Hi Michelle I like that your answer included the term validity coefficient which led me to go back and take a second look at the term which ultimately refers to the relationship between scores on an instrument and the criterion or gold standard measure.Why do you feel like Criterion Validity is common? We were on the same page with the discriminant and convergent evidence validity and used similar examples.

      Reply

    • Christopher LePage
      Jun 04, 2020 @ 13:59:39

      Hi Michelle, I am with Althea, I like how you included the validity coefficients as I am sure it will end up being useful for the exam! We agreed on the fact on why it is so important to have criterion-related variability as well. I think it is extremely useful for mental health professionals to help get a better understanding of their clients, and it helps better ensure the safety of the client as you had mentioned.

      Reply

  3. Yen Pham
    Jun 03, 2020 @ 19:26:04

    1. Discuss your understanding of criterion-related validity (also known as: Prediction or Instrument-Criterion). In you discussion, include why this particular type of validity is common/important for mental health assessments.

    Criterion-related validity, also known as prediction or instrument-criterion is instrument measures how well one measure predicts an outcome for another measure or predicts future behavior. Criterion- related validity is divided into two types: predictive validity, and concurrent validity. Concurrent Validity, there is no time lag (relatively speaking) between when the instrument is given and when the criterion information is gathered. This type used for “immediate prediction”. Predictive Validity, there is a time lag between when the instrument is administered and the time the criterion information is gathered. Criterion-related validity is very important for mental health assessment because it is useful to help a counselor for predicting performance or behavior of clients in another situation (past, present, or future). Basis on that, the counselor will provide a best treatment for each client.

    2. First, discuss the difference between convergent evidence validity and discriminant evidence validity. Second, provide an example of a hypothetical or real assessment where these two types of validity would be important – very common for many mental health assessments.

    Examining an instrument’s relationship with similar constructs and purportedly different variables is called convergent and discriminant evidence. Convergent evidence validity means an instrument is related to other variables to which it should theoretically be positively related. In other words, convergent evidence validity refers to the degree to which two measures of constructs that theoretically should be related are in fact related. For example, if an instrument is designed to measure depression and correlates highly with another instrument that measures depression, then there is convergent evidence. Discriminant evidence validity, on the other hand, exists when an instrument is not correlated with variables from which it should differ. In other words, discriminant evidence validity applies to two dissimilar constructs that are easily differentiated. For example, when an instrument designed to measure depression is not significantly correlated with an instrument that measures trait anxiety.

    Reply

  4. Althea Hermitt McPherson
    Jun 04, 2020 @ 02:15:37

    (1) Discuss your understanding of criterion-related validity (also known as: Prediction or Instrument-Criterion). In your discussion, include why this particular type of validity is common/important for mental health assessments.
    Criterion-related validity refers to the extent or degree to which an overall instrument was related to an outcome or was able to predict a certain outcome in the present or future. Eg of criterion-related validity would be whether excellent SAT or ACT scores or high school grades predict academic performance or achievement in college. Does a certain score on the Colombia Suicide Severity Rating Scale predict present or future suicidality? Mental Health Professional finds criterion-related validity to be extremely important as it allows counselors to get pertinent information that can be used to make predictions about a particular present or future outcome. Therefore instruments can be used to predict clients who are more likely to commit suicide or more likely to succeed at a certain educational attainment.

    (2) First, discuss the difference between convergent evidence validity and discriminant evidence validity. Second, provide an example of a hypothetical or real assessment where these two types of validity would be important – very common for many mental health assessments (hint: listen to my example in the lecture recording
    Convergent evidence validity indicates that an instrument has evidence of validation because the measure is positively related to other instruments that measure the same construct. Therefore if an instrument is originated to measure depression and match up with other instruments that measure depression then we can say that the instrument that was originated is positively related or shows convergent evidence of validity. Any depression scales developed should have a high validity coefficient with similar or well-established scales such as the Beck Depression Inventory.
    On the other hand, Discriminant evidence validity indicates that an instrument has evidence of validation because the instrument is not positively related to instruments that measure different psychological constructs. Therefore a depression instrument should show no sign of correlation to an anxiety instrument because they should be different. There should be a low validity coefficient or validity score because the instruments are different and should not have any close characteristics. For example a depression assessment should not correlate with a Generalized Anxiety scale because they were intended to measure different psychological constructs.

    Reply

    • Christopher LePage
      Jun 04, 2020 @ 13:53:52

      Hi Althea, I like how you specifically added SAT’s predicting academic performance, as it is a clear analogy. I also liked how you mentioned that the test would be helpful to suicidal patients, I am curious as to what other types of “problem behavior” do you think criterion-related validity would also help predict!

      Reply

    • Francesca DePergola
      Jun 04, 2020 @ 16:04:19

      Hi Althea,
      I think your examples of the first questions made sense and made me remember Dr. V’s examples of the false positive and false negatives and the hits and misses. I think it is interesting how whether it is employment or school that you can visualize these things, but a bit scary when things that counselors might look at such as suicidality. It is certainly intimidating if things can be overlooked and misunderstood in areas with such severity, but so important for our field to make sure we understand these assessments.

      Reply

    • Dawn Seiple
      Jun 04, 2020 @ 18:54:08

      Hi Althea,
      In your first section, I think it was helpful to share two distinctly different examples of criterion-related validity, one that predicted academic success and another that focused specifically on a mental health condition. It illustrated the range of tests that require criterion-related validity.
      In the second part of your post, you highlight the theoretical example of convergent evidence when a depression instrument has a high validity correlation with other depression instruments and divergent evidence existing when there is a low validity correlation with anxiety instruments. I cited this example as well, but have wondered how divergent evidence is affected when patients have co-morbid conditions. If a patient suffers from anxiety, does this affect the validity of the tests? Does that make it look like the test is addressing the wrong psychological construct? I am hoping to learn more about this as we get into specific assessments and their uses.

      Reply

    • Selene Anaya
      Jun 05, 2020 @ 14:36:46

      Hi Althea! I think your specific examples of criterion-related validity and their importance really help us to grasp what exactly that type of validity is. I think those examples really represent how important this type of validity information can be, especially when it involves the prediction of possible harmful behaviors or events such as suicide. I never really enjoyed the ACT (I’m sure no one does) and always thought it was not really fair. However, when Dr. V gave us the example in his lecture with the expectancy table and how some colleges use the scores as cutoffs to determine which applications they will actually look at, I can understand it more. Your explanations of both the convergent and discriminant evidence were very clear. Before I read the chapter/listened to the lecture, I never really thought of checking correlations with other assessments to obtain validity information. I think it makes so much sense now and it can really provide us with the information we need to determine whether or not the assessment we are using will test what we want it to.

      Reply

    • Michelle McClure
      Jun 06, 2020 @ 16:55:24

      Hi Althea. I liked the examples you used in your post. I agree mental health professionals especially find criterion-related validity to be extremely important as it gives us important information that can be used to make helpful predictions about possible outcomes for our clients. These instruments can be used to help clients who are depressed and more likely to commit suicide which is extremely important for our clients mental health and safety.

      Reply

  5. Dawn Seiple
    Jun 04, 2020 @ 12:42:28

    1.Discuss your understanding of criterion-related validity (also known as: Prediction or Instrument-Criterion). In your discussion, include why this particular type of validity is common/important for mental health assessments.

    Criterion-related validity tells the user of a given testing instrument how well that instrument predicts a future behavior. If it examines near-term behavior, it is considered concurrent validity. If it examines relatively longer-term behavior, it is called predictive validity. Mental health professionals want to understand how likely it is for a patient to engage in a certain behavior. To decide what testing instrument to use, they need to know the criterion-related validity of the different options. A very valid instrument will more accurately predict future behavior and will allow the practitioner to design the most appropriate treatment plan. This is obviously very important in the example we have discussed in class related to suicide, but applies to any mental health situation where the patient or others could be at risk. Errors in predicting likely future behaviors could result in dire outcomes.

    2.First, discuss the difference between convergent evidence validity and discriminant evidence validity. Second, provide an example of a hypothetical or real assessment where these two types of validity would be important – very common for many mental health assessments (hint: listen to my example in the lecture recording).
    When convergent evidence exists, an instrument designed to test a given variable correlates with other instruments designed to test the same variable. Divergent evidence exists when the instrument does not correlate to instruments designed to test different, unrelated variables. For example, an instrument designed to test depression should have a high validity correlation with other depression inventories. This represents convergent evidence. That same depression inventory instrument should have a low validity correlation to tests designed to assess anxiety. This is considered discriminant evidence. Where mental health disorders are complex, it is helpful to be able to compare different instruments to assess their appropriateness.

    Reply

    • Francesca DePergola
      Jun 04, 2020 @ 13:59:20

      Hi Dawn,
      I like how you mention the suicidality of clients and how is it of paramount importance to have valid instruments in those cases. I have heard others, and even thought this myself about why assessment is important if we are going to be counselors? Well, I am beginning to understand its place in counseling. If counselors are helping clients, they need to be aware of assessments and how to read instruments. In cases like this where anxiety and depression are comorbid in nature, and unfortunately, suicidality being high, it can make a difference between saving someone’s life and ignoring that vital information. I know that sounds a little extreme, but thank you for mentioning that bit of information as it made me ponder more about those instances.

      Reply

    • Casey Cosky
      Jun 06, 2020 @ 16:12:18

      Hi Dawn!

      I really like how you were able to explain multiple types of validity in a clear and understandable way. It’s so important for mental health professionals to be able to predict how a client will react to different types of treatment, and you’re unfortunately completely right that errors in this prediction will not produce a good outcome. It’s honestly kind of scary to think about how an inaccurate prediction can lead to something as harmful as suicide, but that just further proves the point that we need to work as hard as possible to collect accurate data and think critically about different problems and solutions. The more appropriate the treatment plan is for the client, the better chance we have of helping them.

      Reply

    • Michelle McClure
      Jun 06, 2020 @ 17:07:46

      Hi Dawn! I really liked how you mentioned the importance of instruments in treatment planing for clients. I would also like to add that instruments, like those for depression, suicide and self harming behaviors are also very important for creating safety plans for clients which go along with their treatment plans. I also liked your example of convergent evidence using instruments designed to test depression should have a high validity correlation with other depression inventories and also have a very low validity correlation with instruments that score anxiety. I also liked how you end with mental health disorders being complex, isn’t that the truth. The more accurate instruments we have at our disposal the better treatment we can provide our clients.

      Reply

    • Trey Powers
      Jun 07, 2020 @ 15:16:38

      Hey Dawn!

      I found your comments on the importance of criterion-related validity within the mental health community to be very insightful. Prior to this class, I thought that assessments other than simple ones, such as for depression and anxiety, were not particularly relevant to the day-to-day functioning of a counselor, thinking them more useful in research settings. Now, understanding the power these assessments have to predict future outcomes rather than simply understanding where a client is currently, I find myself with a greater appreciation for the use of assessments in counseling.

      Reply

  6. Francesca DePergola
    Jun 04, 2020 @ 13:44:06

    Criterion-related validity, which can also be identified as Prediction or Instrument-Criterion is how well an instrument or assessment is related to an outcome or a predictor of present or future behaviors, for example. Convergent validity would assess the administration of an instrument to the behavior, or whatever the criterion is, without much time in between whereas the predictive validity has lag time within a range. To decipher whether it is concurrent or predictive is up to the administrator’s discretion. The use of this validity is common and important for mental health assessments because it can provide crucial information about the client and their behaviors. If the instruments are used properly, a counselor can assess the results in which provide information if a certain diagnosis, treatment plan, or action is needed to assist their client.

    Convergent evidence validity is when an instrument is related to other variables in which it should theoretically be positively related or similar whereas discriminant evidence should differ between instruments. An example of a hypothetical or real assessment where convergent evidence would be important would be the comparison between two depression instruments. An example of where discriminant evidence would be important would be the comparison between a depression instrument versus an anxiety instrument. Since both instruments are assessing different things they should differ, whereas the two depression instruments should be positively related.

    Reply

    • Madi
      Jun 05, 2020 @ 11:19:53

      Hi Francesca,

      I found your definition to show that you had a clear understanding of criterion-related and convergent validity. I think you brought up a good point that determining if it is criterion-related or convergent is completely subjective. I would want to push you further about how it can help counselors and why the predictive value is important.

      I thought your last sentence for the second question was worded perfectly! It really explained clearly, without fancy words, what convergent evidence is. I found that your first sentence seemed a little wordy and confusing, but I feel as though you became clearer as your response continued. Possibly because by the end of explaining it you understood it better.

      Madi 🙂

      Reply

  7. Christopher LePage
    Jun 04, 2020 @ 13:47:24

    Criterion-related validity is so important, because it helps clinicians be able to predict future behavior. When discussing the importance of Criterion-related validity it is also important to note the two different variations of it: predictive validity and concurrent validity. Concurrent validity is when the tests are administered at about the same time, whereas with predictive validity there is a bit of a break in-between. Criterion-related variability is so important to mental health professionals cause it can help forming a client’s treatment plan. By being able to reliably predict a client’s behavior, clinicians are able to work around the client’s schedule, socioeconomic status, as well as their home, which all play major roles in the treatment process of the client

    The difference between convergent evidence validity and discriminant evidence is what instrument is being used and what for. For convergent evidence validity the instrument and what is being used for match up with one another, giving them a strong level of validity. For instance, if you have a new instrument to measure depression, it should correlate with an already existing depression measure, giving them high validity. With discriminant evidence validity it is the opposite. The two variables are not a match for one another. For example, if you have an instrument of depression, it should not correlate with an instrument designed to measure anxiety, low validity.

    Reply

    • Dawn Seiple
      Jun 04, 2020 @ 18:30:35

      Hi Chris,
      I like the fact that your post reminds us that the validity of a given assessment instrument is just one factor in forming a client’s treatment plan. The practitioner needs to think about the very personal qualities of the patient such as their cultural background and socioeconomic status. With so much focus on data and testing, it can by easy to have the discussion become impersonal. Dr. V eluded to this today, but it is critical to consider the personal qualities of your patient in all aspects of therapy, including assessments, repertoire building and communication.

      Reply

    • Yen Pham
      Jun 05, 2020 @ 14:31:49

      Hi Christopher,

      I like the fact that you have mentioned about the high and low validity (validity coefficient). I agree with you that when we’re looking at convergent evidence, we want to have a high validity coefficient. However, when we’re looking at the discriminant evidence, we want to have a low validity coefficient. If so, both convergent/ discriminant evidence help us to establish construct validity well. The reason for that I think basis on the definition of convergent evidence, which two measures of constructs that theoretically should be related, are in fact related. Thus, we expect a high validity coefficient when we consider using it. Discriminant validity, on the other hands, is demonstrated by evidence that measures of constructs that theoretically should not be highly related to each other are, in fact, not found to be highly correlated to each other. Therefore, we expected a low validity coefficient when we consider using it.

      Reply

    • Casey Cosky
      Jun 06, 2020 @ 16:05:29

      Hi Christopher!
      I like how you pointed out the multiple factors to consider when predicting a client’s behavior. I feel like people tend to focus just on the symptoms without taking schedules, socioeconomic status, and home life into consideration. External factors are just as important to think about when treatment is being planned as symptoms such as negative emotions and suicidal ideation. Being able to predict a client’s future behaviors is so important in order to keep the client safe and also find the method of therapy that fits them best.

      Reply

    • Brigitte Manseau
      Jun 06, 2020 @ 21:40:15

      Hi Chris,
      You raise a great point on the importance of having different criterion-related validity. I now understand how variations of predictive validity and concurrent validity allow flexibility for counselors to better help a client. Prior to reading your post I had not thought about the various external factors that impact the assessment process and forming a client’s treatment plan. It opened my eyes to the importance of considering clients’ schedules and socioeconomic status so that the proper assessment is used.

      Reply

  8. Casey Cosky
    Jun 04, 2020 @ 18:20:00

    Validity is essential to determine how well an instrument measures what you want it to measure and whether it’s accurate in its predictions. Criterion-related validity is common for mental health assessments because it can help predict how a person may act in the future. A test with criterion-related validity should accurately determine outcomes, such as an SAT test predicting how well a student will do in college or a performance test predicting how well an applicant may do at a new job. In the psychology field, criterion-related validity can predict symptoms of mental illnesses and help to highlight certain risks such as suicidal ideation. It’s important because not only does it contribute to a patient’s safety, but it can help someone find the best treatment plan for them.

    In establishing convergent evidence validity, one must show that the measures that are supposed to be related are actually related. To establish discriminant evidence reality, one must show that the measures that are not supposed to be related really are not related. With discriminant evidence validity, it is important to have a low validity coefficient because it is supposed to be different, but with convergent evidence validity there must be a high validity coefficient. A new instrument designed to measure anxiety disorder should correlate highly with a pre-existing instrument that has already been found to measure anxiety disorder. This would be an example of convergent evidence validity. This new instrument, however, should have a low correlation to in instrument designed to measure depression. This is an example of discriminant evidence validity.

    Reply

  9. Trey Powers
    Jun 04, 2020 @ 18:47:49

    1.
    Criterion-related validity, sometimes referred to as predictive or instrument-criterion validity, is an assessment of an instrument’s capability to predict future outcomes based on the instrument as a whole rather than based on its individual parts. Within criterion-related validity also exist two other forms of validity — concurrent and predictive. Concurrent validity produces immediate information about the results. Predictive validity is used together with concurrent validity to make a prediction of what future results will look like (e.g., making a prediction of how a person will be feeling/acting in the future for diagnostic purposes). Criterion-related validity is particularly helpful in the mental health field because clinicians need to know that the instrument being used is appropriate for identifying a certain disorder or condition, and also be able to have an idea of the disorder’s progression over a certain period of time. Choosing an appropriate measure can therefore be a matter of life or death potentially, so it is essential that clinicians have examined the validity of the instrument they have selected and determined that it is applicable in the given situation.

    2.
    Convergent evidence validity indicates that the instrument assessing a particular construct is similar to other instruments that assess similar constructs. If there is high validity, it indicates that the instruments are similar, whereas if there is low validity, it indicates that the instrument you are using is not properly assessing the construct, as it should theoretically align with the results of other similar instruments. In contrast, divergent evidence validity indicates that the instrument assessing a particular construct is not similar to other instruments that assess different constructs. In this case, low validity is desired, as the two instruments are not attempting to measure the same construct, and therefore should not have similar results. As an example, if someone was to compare two well-documented depression instruments, we would hope to see that there exists a high level of convergent evidence validity, as they are both intending on measuring depression. On the other hand, should you be comparing a depression instrument with an instrument measuring mania, you would want to see a low level of validity as the two instruments are measuring very different disorders.

    Reply

    • Selene Anaya
      Jun 05, 2020 @ 14:45:25

      Hi Trey! I think your clarification about criterion-related validity assessing the instruments capability to predict future outcomes is based on the instrument as a whole rather than its individual parts was an important distinction to make. I think I forgot to include that in my definition. It is the one key distinction between criterion and content-related validity. Content related concerns the individual parts of the instrument. I think the fact that you mention progression instead of solely predicting future outcomes was interesting as well. I wonder if there is a touchy distinction between the two. Your examples explaining both convergent and discriminant evidence were helpful in understanding both concepts as well. I said it before, but I never really thought of this being a good way to gather validity information about assessments, but it makes total sense now after lectures, our textbook, and these discussions.

      Reply

  10. Selene Anaya
    Jun 04, 2020 @ 21:36:37

    1. My understanding of criterion-related validity, is that it tells us the extent to which a specific criterion can be predicted. An example of this type of validity would be how well ACT and SAT scores predict success in college. This type of validity is especially important for mental health assessments because in the mental health field, we mostly use assessments to either assess or predict an individuals behavior, emotions, or cognition. Therefore, we want to obtain validation evidence that will predict the future outcomes of specific criteria relevant to what we are assessing. Given that, we are relying on these assessments to give us statistical data that will allow us to gain a reliable (and valid) prediction of our clients future thoughts, feelings, or behaviors. This type of validity can inform us about the extent to which the important behaviors we are assessing can be predicted so we can determine a treatment plan we want to pursue with our clients. There are also two types of criterion-related validity. Concurrent and predictive with the main difference being time between taking the instrument and obtaining the criterion (predictive) information. Concurrent will give us information right away such as when we want to diagnose. Predictive validity has more time between. Both, however, have a judgment component meaning it is up to the authors to decide on the time period between concurrent and predictive validity. It could be 2 weeks to 10 years, or 1 day to 5 months. An example of criterion-related validity, making it clear how important this type of validity can be, is if a client is given a suicide assessment. We want to first ensure it is appropriate for the individual, and then make sure it will inform us with the most accurate prediction if the individual will engage in harmful behavior in the future. From there, we can take the steps necessary to act.

    2. When we look at validity, it is helpful to find evidence supporting the instrument based on relations to other variables to add to our knowledge about the instrument. One of the most common ways to do this is to look at how it correlates with other assessments. There are two types of evidence validity that we can look at. Convergent evidence validity is when we compare an instrument to other instruments that are either more developed or are within the same domain. With convergent evidence validity, we want a high correlation coefficient to exist between the two instruments because that will tell us how well they compare and that it really tests what it is intended to test. An example could be designing an instrument to assess anxiety and comparing it to another instrument that also measures anxiety. If the correlation is high, then we would have convergent evidence assuring our anxiety instrument measures what it is supposed to. If it is low, it might indicate there are some flaws in our created assessment. Discriminate evidence validity can also be extremely helpful. This type of validity is the extent to which an instrument does not correlate with another from which it should differ. Using the previous example of our anxiety assessment, if we correlate it with a well established assessment for depression, in order to obtain discriminate evidence, we want a low correlation coefficient because that would determine that it discriminates between assessments. If the correlation was high, this wouldn’t be good because that means our assessment to test anxiety, could or does also assess depression, and that isn’t what it is intended to measure.

    Reply

    • Brigitte Manseau
      Jun 06, 2020 @ 21:06:18

      Hi Selene,
      I like how you explained that the authors have the freedom to decide what the time frame is for concurrent and predictive validity. I initially found it mind-boggling how big of a gap there can be when either concurrent or predictive validity is compared between similar instruments. From the lecture and reading the book, I realized how important it is to remember that all instruments vary in terms of their individual concurrent validity and predictive validity. In terms of the second prompt, you clearly described what high and low correlations in convergent evidence and discriminant evidence meant. I found it very helpful because you highlighted what correlations are *not* wanted. I tend to grasp material better when the whole picture is explained rather than just the part that may be deemed important.

      Reply

  11. Brigitte Manseau
    Jun 04, 2020 @ 23:58:52

    1. Criterion-related validity measures how well an instrument predicts certain future behaviors. If the instrument predicts the behavior “now” meaning within a week or month it is considered concurrent validity. On the other hand, if the instrument predicts the behavior farther off in the future such as eight months from now it is considered predictive validity. Criterion-related validity indicates whether an assessment counselors use is a good predictor for the future behavior of clients. For instance, it is important for a counselor to use an instrument that predicts future harmful behaviors of a client who is suicidal.

    2. Convergent evidence shows when an instrument is related to other variables which should be positively related hypothetically. Discriminant evidence is when an instrument is not correlated with other variables that it should differ from. Convergent evidence compares two similar instruments while discriminate evidence compares two instruments that differ from one another. For instance, if I created an anxiety instrument I would expect convergent evidence when comparing it to an existing well established anxiety instrument. Comparing it to another reliable and valid anxiety instrument would help indicate whether my instrument measures anxiety well. A high validity coefficient would indicate convergent evidence. Discriminant evidence would be expected when the anxiety instrument is compared to a well established depression instrument. Depression and anxiety instruments are good to compare in this instance since they have symptoms that over lap each other. Therefore, it is super important the instrument measures just anxiety and not depression. A low correlation coefficient would indicate discriminant evidence.

    Reply

    • Madi
      Jun 05, 2020 @ 11:24:33

      1. I thought you showed a clear understanding of the topic! I liked how you used the term “now” showing that it is very subjective determining if it is criterion-related or concurrent. I also loved your example of how this is important for counselors. I thought it really showed how important this is.

      2. You see to really understand this stuff! I liked how you explained how convergent is two similar instruments and discriminant is two instruments that are not similar. I also think it bringing in the coefficients shows that you have a strong understanding of these concepts!

      Madi 🙂

      Reply

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

Adam M. Volungis, PhD, LMHC

Enter your email address to subscribe to this blog and receive notifications of new posts by email.

Join 58 other followers

%d bloggers like this: