Topic 1: Norms and Meaning of Test Scores {by 8/27}

Based on the text readings and lecture recording due this week consider the following two discussion points: (1) In your own words, provide a general description of the difference between criterion/domain-referenced instruments and norm-referenced instruments.  Why is it important to understand this difference?  (2) What are your thoughts on your understanding of the normal distribution?  In other words, does it generally make sense (explain) or is it still a little confusing (that’s okay – explain)?

 

Your original post should be posted by 8/27.  Post your two replies no later than 8/29.  *Please remember to click the “reply” button when posting a reply.  This makes it easier for the reader to follow the blog postings.

77 Comments (+add yours?)

  1. Elias Pinto-Hernandez
    Aug 25, 2020 @ 09:58:46

    Criterion/domain-referenced tests are instruments that measure the performance of each candidate evaluated by itself unrelated to a group. On the other hand, norm-referenced instruments evaluate the participant’s position with respect to others in his group. Criterion and norm tests are tools to gather data; without data, we have no way of knowing if our hypothesis is correct. In another way, it is just a belief, not science. It is imperative to understand the difference between the instruments because they measure different data.

    For a normal distribution, I understand that it is a bell-shaped and symmetrical distribution where successive standard deviations from the mean establish reference values for estimating the percentage of observations in the data. It is a model that approximates the value of a continuous random variable to an ideal situation. The reference values are the basis of many hypothesis tests. The normal distribution can be used as a tool to calculate probabilities.

    Reply

    • Beth Martin
      Aug 25, 2020 @ 19:54:54

      Hi Elias!
      I really like your description of the model of normal distribution of “continuous” – it makes it much easier for me to picture the measurement as a sliding scale, rather than set points (if that makes sense). I also agree that it’s important to be aware of how different the things the data for both instruments can represent, especially when we’re looking at behaviors that can be seen as different from the norm!

      Reply

    • Abby Robinson
      Aug 25, 2020 @ 22:30:50

      Hi Elias, I really liked how in your understanding of normal distribution you say it is a good way to predict probabilities. I never thought about how this might be useful in that sense but I can see how you could use it because of generating the average or norm of a group of scores. In this sense you could say that if the average score of a test way an 80%, more tests will also be an 80% because that is how the group of scores was analyzed.

      Reply

    • Tayler
      Aug 26, 2020 @ 11:43:23

      Hi Elias! I like your discussion of criterion-referenced tests. But, couldn’t you argue that both are ways to gather data? I would think that data from individuals, creating a norm, might even be more effective. After all, if a test is way too difficult or way too easy, how will you know what a criterion-referenced test score means?

      Reply

  2. Tayler Weathers
    Aug 25, 2020 @ 12:56:50

    1. The main difference between criterion/domain-referenced instruments and norm-referenced instruments is whether or not the score is compared to other individuals. Criterion/domain-referenced instruments focus on an individual’s performance within the test, i.e., did they get 70% of the questions correct or not. This has no indication of how difficult the test might have been to others; it just tells you how they did. Norm-referenced instruments, in contrast, focus on an individual’s performance compared to others. So, a 70% might seem like it was good, but not if every other person taking the test got a 90% or higher. It’s important to understand the difference because it tells you how to interpret the scores – for a test like the SAT, it’s better to get a higher score, not just because it indicates you know the material, but it also indicates that you know the material better than others taking that test. This also gives some wiggle room for a particularly difficult or easy test; if everyone gets a really low score, that will be accounted for (kind of like a curve) in the final score report.
    2. The normal distribution makes sense to me! I feel confident with the mean/median/mode from various statistics courses I took in undergrad and before.

    Reply

    • Beth Martin
      Aug 25, 2020 @ 19:59:40

      Hi Tayler!
      Your example of what test scores in both instruments look like was really helpful – I didn’t grow up with norm-based instruments in school (which I understand American children do have to get used to with SATs and curves etc), and in my undergrad stats classes I would get beyond lost! This was a really simple, easy-to-follow explanation. I hadn’t really considered that norm-based instruments can reflect difficult tests (50% is still 50% if all of the sample struggled, rather than everyone failing), which is definitely making me wish I’d grown up in a system that does this!

      Reply

    • Abby Robinson
      Aug 25, 2020 @ 22:24:36

      Hi Tayler, I really liked that in your definition of criterion referenced instruments, you mention that the test doesn’t show how hard it is, just whether you know the information or not. I think this is a great way to think of it because there is only one baseline really. Whereas, when you explain the scores in the norm referenced instruments, you may have gotten a 70% on your exam where as another person may have gotten a 90%. It’s a good way to show that the average test score really shows how the group scored overall! The SAT is a great test to show as an example because there is a large number of people taking that test and your score is very much compared to how the group did as a whole!

      Reply

    • Bibi
      Aug 27, 2020 @ 17:32:52

      Hey tayler!
      I really liked how you started off your definition by clearly defining the difference between norm and criterion referenced. Then the extra examples you gave were really helpful in making sure that I knew how they were explained. It was super helpful to read through

      Reply

  3. Beth Martin
    Aug 25, 2020 @ 20:00:26

    1) A criterion/domain-referenced instrument is an instrument measures performance based on what an individual knows, rather than by assessing performance by comparing them to others. Criterion/domain-referenced instruments have set boundaries or markers that indicate whether an individual has met a standard, e.g. whether a student has understood enough material to pass an exam. The instrument does not take into account the scores of other students to determine if this particular individual has passed. If the student has hit a certain score (such as a 70% grade), they have passed the test, regardless of the scores others have achieved.

    Norm-referenced instruments, however, are based on how an individual performs in comparison to a group that have previously undertaken a test or exercise. Their scores are converted into figures/measures that allow for comparison across tests and within groups. SAT percentiles, for example, measure how well an individual has done in comparison to other people in their group – a 70% rank indicates that a student has scores higher than 69% of their peer group, as opposed to correctly answering 70% of questions (as in criterion/domain-referenced instruments).

    It is important to understand the difference between the two as they can infer very different things. A raw score obtained from criterion/domain-referenced instruments means than an individual has passed (or failed) certain markers that professionals deem important to this particular area. Scores obtained from norm-based instruments, however, measure how an individual compares to other people, not necessarily their own, concrete, understanding. In regards to psychological measurement, behaviors may be compared to the general population norm, or against set criterion (e.g. anxiety levels vs neurological functioning). Knowing the difference between norm-referenced instruments and criterion/domain-referenced instruments will allow a clinician to better select assessment tools, and understand the data that they get back with that added context.

    2) Normal distribution does make sense to me! It seems sensible that mean is used to base distribution off of, as that’s the “average” standard of a sample size, and that that’s what we base everything else around. The less “average” a score is, the rarer it is, so the curve itself is a good way of visualizing it (for me, anyway). Percentile ranks were something very new to me (we don’t really use them in education where I’m from), but I think I’ve gotten to grips with it and standard deviation.

    Reply

    • Tayler
      Aug 26, 2020 @ 11:47:44

      Hi Beth!
      I think it’s interesting that you focus on the distinction’s importance in terms of the mastery component! That is something I feel I forget a lot, and I think it’s important when trying to envision a person’s understanding of something. After all, just because you understand something more than 70% of other people doesn’t mean much – especially if all of those people just walked in to the test cold! I used to teach SAT prep classes, and the idea of a norm-based test was something I really emphasized to my students so that they didn’t get bogged down in the “well I need a perfect score because I’m smart!” mentality. So it’s good to remember that the criterion-reference tests are useful, even if I’m not as used to them.

      Reply

    • Viviana
      Aug 26, 2020 @ 22:14:30

      It took me a little while to understand percentile ranks too. I believe the term is confusing to a general population. Interpreting percentile ranks could be confusing and somehow I need to make sure I don’t confuse it with percentage. For example, it is easy to misunderstand a client’s assessment result categorized in the 40th percentile or a score of 40%. As you said the visualization of these results are easier to read in a visually and the bell curve allows it.

      Reply

      • Lilly Brochu
        Aug 27, 2020 @ 08:27:52

        Hi Viviana,

        I also get a little confused about the terms, but I hope that with practice and by familiarizing ourselves with this material, it will get easier in time. In my opinion, one of the most confusing parts of the PowerPoint lecture was to understand the exact percentage an individual falls into above or below the standard deviation. Maybe I am just a little rusty with my math skills, but I was struggling to understand how our professor was getting certain percentages. I have not taken a statistics course since sophomore year of undergrad nor any math courses since. I do not think I am alone in saying that some of us may need a bit of time adjusting to the math portion of psychology!

        Reply

        • Maya Lopez
          Aug 29, 2020 @ 21:04:59

          Hi Lilly and Viviana,

          I agree the terms percentile rank and percentage are quite easy to get mixed up. I think that was when the Professor was explaining how a 1st grader can be reading at a 4th grade level,but they never even gave the kid 4th grade reading material; it was just the extrapolation of data. I also struggled when the Professor was calculating SD and other percentages but mostly because I am a visual learner and got lost when so many numbers were added and multiplied at once. You are certainly not alone about the math! I am a bit rusty on definitions and how to calculate certain things. Hopefully the next lecture goes better for everyone!

          Reply

    • Bibi
      Aug 28, 2020 @ 13:17:42

      Hi Beth. I really like your detailed descriptions of what a norm referenced with criterion referenced test is. I felt like you gave really good definitions that helped clear up any misunderstanding I had about the two (I actually out your definitions in my notes). Also the explanation of why it is important to understand the difference between the two because of what they are used to measure was super helpful because I felt like it really added to when you would want to use each type of test.

      Reply

  4. Bibi
    Aug 25, 2020 @ 21:53:22

    1. Norm referenced means that the score of any given individual is compared to the scores of other people who have taken it. There is a population that is used for the norming group. They take the measure and then their scores are used to create a standard. Then when you take the measure, your score is then compared to that of people in the normed group
    When a measure is criterion referenced then your score is compared to an established standard. For example, there is an established standard for the ACT exam that your score is compared to and then your grade is determined based on that standard.
    2. A normal distribution means that a majority of the scores fall along the mean, median, and mode. This creates a bell curve with extreme scores on the ends of the curve with fewer frequencies. I think the normal curve makes sense overall. It is something that we spent a lot of time talking about in undergrad stats classes.

    Reply

    • Elias Pinto-Hernandez
      Aug 29, 2020 @ 16:10:33

      Hi Bibi,
      Thank you for your pots. I appreciated your description of this week’s topic. Your example made sense to me, especially the normal distribution.
      Excellent!

      Reply

    • Maya Lopez
      Aug 29, 2020 @ 20:53:46

      Hi Bibi,
      I really liked the way you described a normal distribution curve, it’s a good thing you spent a lot of time on it in your undergrad stats class. It makes sense that the most common scores along the average would be in the middle and more extreme ones on the end. Lastly, I wonder if people who are chosen to be in a norming group, know that they will be helping to set the standard.

      Reply

  5. Abby Robinson
    Aug 25, 2020 @ 22:16:59

    In my own words, the definition of a criterion/domain-referenced instrument is that this test is to evaluate what the individual knows/comprehends. There is a “mastery” level that if achieved, the individual knows the information, they’ve passed. For example, if you had to take the BAR exam to become a lawyer and you needed an 85 to become a lawyer, if you’ve submitted your test and you get any score 85 and above, you’ve passed and you’re now a lawyer. The mastery level is predetermined before the test begins. This test score is not compared with any other scores taken by others. In my own words, the definition of norm-referenced instruments is that they are tests that take the score of an individual and compares it to a group of scores (they have taken the same test). The group that has taken the same test as you that your score is being compared to, is called the standardized sample. By comparing an individual’s score with the sample scores, it gives the results on how the individual stands compared to the rest of the group. The difference between criterion/domain-referenced instruments and norm-referenced instruments is that criterion instruments don’t compare an individual’s score with any other scores from that test. Also, in criterion-referenced instruments there is already a level of mastery saying that the individual’s score either passed or failed that exam. Whereas norm-referenced instruments do not have a pass/fail aspect, but rather it compares the individual’s score to the standardized sample score. This can show on a scale how the individual did. This is important to understand because there are different instruments that should be given in each situation. It is important to understand the appropriate time to use each instrument and that if a certain type of test is used inappropriately, the scores and results may not have any significance to the case.
    Normal distribution makes sense in the fact that it is used to show the results of many scores from a test. I think it is important because it gives a ‘base’ to compare the results to and gives us a spot to put the information into. Once the scores are calculated accordingly (mean, median, mode, etc.) we can then calculate how to put those scores into useful information that will be relevant to either the public, or client. I think that normal distribution helps put the scores and their information into significant terms that can be understood by most people. For example, if I just told my client that he/she received a 55 on their test, that doesn’t give them any helpful information. But, if I were to convert the score into a z-score or t-score and use that information in normal distribution I could then say to my client, ‘your test results show you are in the ___ percentile.’ Or I could say that their score compared similarly or differently than the norm, which would be more useful than just their original test score.

    Reply

    • Viviana
      Aug 26, 2020 @ 21:46:47

      Abby, you bring up an important point of making these results into meaningful terms to clients, parents, or other collaterals. Often times assessment results are not communicated effectively and individuals get confused and not making sense of the counseling direction. Counselors need to be knowledgeable about the instruments and connect the results to the client’s needs and skills.

      Reply

    • Tanya Nair
      Aug 28, 2020 @ 19:31:44

      Thank you for your comment, Abby. I enjoyed the examples you used and how you connected them to better reference the meanings of criterion and norm-referenced instruments. I think it is important how you used mastery as a term to understand and explain criterion-referenced instruments. It is something I often leave out when I make my own understanding of this. I also liked how you used z-scores and t-scores to enhance your example and show that it is more useful when it is converted into one of these scores as it shows the percentile. I also like how you emphasized when to use which measurement and showed why this is important. I brought this up in my comment as well.

      Reply

    • Brianna Walls
      Aug 29, 2020 @ 15:42:19

      Hi Abby! I really liked how you gave an example of a criterion/domain-referenced instrument, this made it clear to me of what types of exams are used in real world settings. I also liked how you made it clear that a norm-referenced instruments do not have a pass/fail aspect. I also like how you explained how maybe a client wouldn’t understand completely what their test scores mean without being able to covert them into z scores or a percentile. This makes it easier for the client to understand how they performed on a certain exam.

      Reply

  6. Carly Moris
    Aug 26, 2020 @ 14:07:10

    Criterion-referenced instruments compare how a person does on a test to a pre-set standard. These tests can be used to tell how much a person knows about something, or if they have reached an adequate level of knowledge in a certain area. These types of tests just tell you how much a person knows about the subject, not how their performance compares to others. Norm-referenced instruments tell you how a persons performance compare to other people’s performance on the same test. But they do not tell you how much a person knows about a certain subject. It is important to understand the difference because because the instruments use the individuals score in a different way so the results will need to be interpreted differently.

    My understanding of a normal distribution is that the data will be distributed along a bell-curve. In a bell curve the mean, median, and mode will be the same and the majority of the data will fall within one standard deviation above and below the mean. A bell curve has a peak in the middle and the data is distributed symmetrically above and below the mean. In a normal distribution we can use standard deviations to interpret how a score relates to the mean. The greater the standard deviation the greater the score differs from the norm. The smaller the standard deviation the closer the score is to the norm. By figuring out a client’s z-score (the clients score minus the mean, then divided by the standard deviation) we can find out the clients exact percentile score by looking at a conversion table. This is helpful because it tells us what percent of people scored at or bellow their score.

    Reply

    • Pawel Zawistowski
      Aug 28, 2020 @ 13:44:49

      Hi Carly! I really appreciate your detailed description of the normal distribution and that you even touched on z-scores and that we are able to figure out where a person’s percentile may fall with the use of z-scores and the bell-curve. Awesome explanation!

      Reply

  7. Viviana
    Aug 26, 2020 @ 17:49:12

    (1) In your own words, provide a general description of the difference between criterion/domain-referenced instruments and norm-referenced instruments.  Why is it important to understand this difference? 

    Criterion/domain or also called objective-referenced instruments measures what an individual knows and doesn’t know at the time of the assessment; these skills being measured are already stablished at some standard and are not compared to another’s individual’s performance while norm-reference measures the individual’s performance in comparison to the knowledge of another individual’s performance in the same assessment. It important to understand this difference between the two instruments as it’s difficult to tell the difference just by looking at them as the the difference is actually in the scores and it should be interpreted differently to offer an accurate interpretation.

    (2) What are your thoughts on your understanding of the normal distribution?  In other words, does it generally make sense (explain) or is it still a little confusing (that’s okay – explain)?

    My understanding of normal distribution is that it shows that data near the mean are more frequent in occurrence than data far from the mean. The bell curve is a good way to visualize it as it would show the less standard results at the end of the curve.

    Reply

    • Timothy Cody
      Aug 28, 2020 @ 21:00:52

      Hi Viviana! Well thought out response for the first question. you clearly understand the difference between the criterion-referenced instruments and norm-referenced instruments. My suggestion for the normal distribution is to know the cases in which it is used for, especially when using a z-score and t-score. This is when you can compare one’s score to the mean and identify the percentile based on the norm curve. Other than that, well done!

      Reply

  8. Lilly Brochu
    Aug 26, 2020 @ 20:15:08

    A criterion/domain-referenced instrument is used to compare one’s performance on an instrument to a specific criterion or established standard. It does not test an individual on how well they compare to others taking the same instrument. This instrument can portray how knowledgeable one is in a specific area. Additionally, the criterion/domain-referenced instrument has a cut off score (mastery) that one must obtain to pass the test. This score is already decided prior to the test being taken. Determining the mastery can be difficult because it is hard to figure out what and how much should be included in a mastery score. However, the domain/criterion-referenced instrument can be problematic and tricky to figure out the level of mastery one should have in that content.

    On the contrary, a norm-referenced instrument is based on a comparison of an individual’s score with other people’s scores who have both taken the same instrument. The “norming group”, or the group of people who have taken the same instrument as that individual typically vary in size as it could be a very small or large group. Typically, a larger group is better to provide a more accurate comparison. It is important to look at the size, type of sampling (simple random, stratified, or cluster), and how representative the norming group is when interpreting a norm-referenced instrument. It is important to know the difference between the two instruments, so that when assessments are given, there is actual significance to the results that are found. Additionally, the better we know our instruments, the more accurate our future assessments will be
    .
    Overall, when I was reading through the lecture notes and listening to the PowerPoint you posted, it jogged my memory a bit about normal distributions and what I had learned previously in my Psychological Statistics course in undergrad. The normal distribution has a bell-shaped curve and is symmetrical. The middle of the curve represents the mean, mode, and median of the data. As the bell-curve extends to the left or right, it accounts for those who fall below or above the average. It has been a while since I have seen all of the “nitty-gritty” statistics terms, but I am sure that I will be able to recognize and learn them pretty quickly through our more in-depth lecture on Zoom tomorrow.

    Reply

    • Lina Boothby-Zapata
      Aug 26, 2020 @ 22:00:31

      Hi Lilly,
      I agree with you about the importance of the difference between Norm-Reference Instruments and Criterion- Reference Instruments, recognizing that the better we know more accurate the assessment will be, I also think that the better we know our instruments, the communication with our clients is clear and assertive and we can incorporate these results throughout the therapeutic process.

      Reply

    • Christina DeMalia
      Aug 27, 2020 @ 16:54:35

      Hi Lilly, I think it is great that you mentioned the way norming groups work, and how some can be more accurate than others. I think it could be easy to confuse a percentile compared to a norming group, with a percentile compared to every single person who has taken the assessment. Scoring in the 75th percentile doesn’t necessarily mean you scored higher than 75% of the people who have ever taken the exam. Instead, it means you scored higher than 75% of the norming group. Ideally, this group will be large enough and diverse enough to represent the the whole population of people taking that assessment. I definitely agree that the norming group used is important to be aware of so you can be sure the results are accurate and understand where the base line or mean is coming from.

      Reply

    • Nicole Giannetto
      Aug 31, 2020 @ 14:31:08

      Hi Lilly! I enjoyed your example of measuring knowledge or intelligence using criterion/domain instruments I think that incorporating a real life example to theories, definitions and research. I also agree with you that reading about and seeing examples of normal distribution along the bell curve definitely jogs my memory which shows how much practice can improve our understanding of things.

      Reply

  9. Tanya Nair
    Aug 26, 2020 @ 21:06:24

    Norm-referenced measurements and criterion-referenced measurements are certainly very similar in their approaches however, there are some differences. Criterion-referenced measurements use some sort of standard, or any other criteria to assess how well a student is doing. This approach to measurement does not take into account how other students are doing in comparison and instead uses its own “cut-off” scores such as average and proficient to place students into various categories. Therefore, how other students perform does not have any interaction with an individual’s score. Norm-referenced measurements compare how an individual is doing in comparison to his or her peers. Depending on the assessment, the norm group is usually representative of multiple students to obtain the right performance levels. This approach differs from criterion in that individuals are not placed into categories on how well they perform but get a percentile rank according to how they performed in comparison to students of their age.

    I do not think there is one type of measurement that fits all. I think that it is important to understand the difference between these two because according to what is being measured either norm or criterion can be more powerful. This is the case because a student may be labeled as “typical” according to their norm-referenced measures however, not achieve proficiency according to criterion-referenced measures. If I had to choose one though, it would be criterion because the criteria for certain categories can always be changed due to performance from individuals. In some way, criterion measures do take into account norm measures to get the criterion for different categories they need to assess how people are doing first. Whereas, norm measurement is solely based on the performance of others and some students have different developmental rates especially when they are younger.

    The normal distribution makes sense to me as I have learned about it in previous courses. I am pretty familiar with the normal distribution which refers to a bell-shaped curve where most of the data points are somewhere towards the middle and not leaning to either side which makes it positively or negatively skewed.

    Reply

    • Lina Boothby-Zapata
      Aug 26, 2020 @ 22:18:19

      Hi Tanya,
      Based on your comments, It seems to be that the Criterion-Reference instruments do not label the individual and every time that they present the test; such as GRE they have the opportunity to pass and obtain the level of mastery that it is required. I like this observation that you made. Contrary to the Norm-Reference Instrument labeled the client; such as “Typical”. My thoughts are that the benefits of the Norm-Reference instrument are the statistics and mathetics that were introduced to do the instruments, these tests are accurate and objective. However, the only benefit that I see is that the client is being compared with another group. I am wondering how we can use this information in counseling for our clients and how we can present it in a way that is beneficial but no competitive. I am also wondering if there are other benefits of the Norm Reference Instruments.

      Thanks for your comments 🙂

      Reply

      • Lina Boothby-Zapata
        Aug 27, 2020 @ 09:20:17

        In my words, a criterion-referenced instrument is also called a domain-reference instrument or objective-reference instrument. The criterion/domain instrument is characterized by testing specific domains. Instruments such as a GRE test, Social Work ASWB, or the LMHC test are designed to test your knowledge of a particular field. Another characteristic is that these instruments are not compared with a sample group. In summary, the instrument test your mastery, meaning your knowledge. However, Criterion/domain reference instruments have limitations. The first limitation is about how they measure this domain, what the content is that they are going to measure, and how they select the content that is going to be assessed. The second limitation is related to how to determine the score that defines that the person has master the domain that is being tested or if the individual has accomplished the expected performance on the test. As an example, one of my own experiences is that DCF agency requires all social workers to be LSW licensed. Now, the agency assumes that test is measurement knowledge in your field, child protection. However, when you present the test, there are no questions related to this domain at all; colleagues recommended to others, “If you want to pass the test, don’t answer as a DCF worker.” I guess the ASWB Board of Association needs to develop a new instrument to assess the domain of Child Protection Social Workers. Contrary, Norm-Referenced Instruments are designed to compare your score with the rest of the group that also performed the test. This group is called the norm group or standardize sample. Also, with the Norm- Reference Instruments statistics and mathematics has been incorporated to explain the results and the comparison between the individual and the group, and then it really gets complicated.

        In summary, the main differences between Criterion-domain reference Instruments and Norm-Reference Instruments are that Criterion Instruments assess based on the domain or area that the instrument is testing, while the Norm Reference tells you how you did in comparison or in relations with the group. Furthermore, Norm-Reference Instruments incorporate statistics meaning that the results will provide specifics about his/her performance and objective and accurate interpretation and, as a consequence, credibility for the clients on the results. Contrary to Criterion-Domain instruments have been challenged because they have questions about how to they select the information to be assessed and how was the standard score determinate, and they don’t support their results on statistics. Now, it is crucial to understand these differences because as a counselor, I will be in the process of assessing a client and selecting the instrument to apply in a responsible matter.

        The way that I understand the normal distribution is that there is a statistical form to organize the information, in this case, the results of the instrument. After your information is collected and classified, then the Standard Deviation can be applied and provide more specific and accurate information about the results. Now, the normal distribution is the “normal graph” that allow us to convert and visualize the standard deviation, the Normal curve has mathematical properties that make it very useful in the interpretation of Norm-Reference Instruments.

        I guess what it gets into my mind is how all these theories about reliability and validity can be translated to our clients in a way that they can understand the purpose, the process, and the results of what we are doing. I would also like to test the instruments in the clinical practice and measure the impact to our clients in the counseling process.

        Reply

        • Lilly Brochu
          Aug 27, 2020 @ 13:37:13

          Hi Lina,
          Thank you for including the limitations of each of these instruments. I think it is important to note the limitations of an instrument as well as its strengths. Furthermore, using your own personal experiences to relate to the material we learn is also a great way to fully understand the content. By comparing the similarities and differences between each of the instruments, we are able to understand and assess our clients more efficiently, and know which instruments are useful in what contexts. I have definitely overlooked the assessment process, and how important it is prior to learning more about them. I look forward to using them in the future with my own clients as well.

          Reply

        • Anna Lindgren
          Aug 29, 2020 @ 14:08:39

          Hi Lina,
          Thank you for not only defining these terms, but providing your own experience of the limitations of these types of instruments. As we move from the theoretical to the practical applications for these types of assessments in our clinical work, insights like these will prove really helpful!
          In your explanation of the normal curve, you say that it is useful of the interpretation of norm-referenced instruments. I am wondering if the normal curve can also be used to show data from a criterion-referenced data set. It may not be as typical, but I think it can be done (i.e. showing the test scores of a final exam in a class). Although, perhaps there would be more instances of skewed data when it comes to criterion-referenced instruments. What do you think?

          Reply

  10. Pawel Zawistowski
    Aug 26, 2020 @ 22:59:41

    1. A norm-referenced test measures what is normal performance for a certain population of students at a given task. For example, how well 12th grade High School students perform on the SATs. If a student scores within the 90th percentile, it means that they have performed better on the given test than 90% of 12th grade students that have taken the same test. Criterion-referenced tests measure performance based on if a student has met a specific criterion. They may have a cutoff score which indicates whether or not the student is proficient at the subject that they are being tested on. For example, a student who is taking a first-aid class may have to score 80% on a written test in order to achieve a passing score to demonstrate that they have acquired the knowledge to certify them in first-aid training. A criterion-referenced tests is more individualized in the sense you must achieve is specific score in order to qualify as opposed to being compared to how well you perform compared to other students. It is important to know the difference because they both have a very different yet important purpose in standardized testing. Norm-referenced tests are important because they evaluate and compare a student’s performance to their counterparts and provide useful information of what is the “norm” for a student. Whereas criterion-based performance is used to assure that a student has acquired required knowledge based on the required criterion that is necessary to make them proficient at the subject at hand.
    2. A normal distribution generally makes sense to me. In life, almost anything you take measurement of will create a bell curve. For example, if you take the height of people at Assumption University. You will have a mean which will represent a majority of students and you will have students who deviate from the mean at a smaller percentile in both directions; taller and shorter. Then, you will also have students on the extreme tail of the distribution who are very tall but also students who are much shorter than the average student. A normal distribution provides useful information what is the norm and the way things may deviate from the norm.

    Reply

    • Zoe DiPinto
      Aug 27, 2020 @ 16:37:21

      Hey Pawel! I also used the SAT to help define norm-referenced instruments. My thought process was similar in that I thought about student’s percentile rank being an essential part of their score. A raw SAT score won’t determine whether a student gets accepted into their dream school. Instead, it’s their score compared to the average score of the other applicants that year. It makes me wonder how the averages of SAT scores have changed over the past 50 years and how we as psychologists should think about the implications of time and score expectations over years. I would think as time goes on, older generations have higher and higher expectations of the scores of younger generations. But maybe someone could argue the other way around?

      Reply

  11. Cailee Norton
    Aug 27, 2020 @ 13:41:54

    1. A criterion-referenced instrument examines how an individual compares to a set standard or domain. This instrument is focused only on that individuals abilities in comparison to that standard. The difference between this instrument and a norm-referenced instrument is that a norm-referenced instrument is used to compare to a group of other individuals. It is not focused on that individuals abilities, as is the criterion-referenced instrument, but only on how that person compares to a group of people. It’s important for counselors to understand these different instruments in order to interpret the results with the client. Being able to read the results back to the client with full understand of who the group of comparison is as well as what the specific domain being test is, specifically in the case of criterion-referenced instruments, allows accurate assessment and discussion with the client.

    2. I feel like I understand normal distribution fairly well. In general term it does make sense, but it never hurts to have a refresher on the subject especially in using it in application to material. Basically, a normal distribution is that the normal curve resembles that of a bell shape as it represents the largest number of frequencies. In order for a distribution to be considered normal there must be a single peak at the center that reflects a mean, mode, and median that are all the same score. When a distribution is not normal it is either negatively skewed or positively skewed, in which a majority of the scores are on either end of the distribution.

    Reply

    • Zoe DiPinto
      Aug 27, 2020 @ 16:29:13

      Hi Cailee, I think your description of a normal distribution is very clear. In my head, I know I understand what a normal distribution is, but I still found it hard to put into words. Your explanation put a clear picture in my head of a bell curve as well as where the frequencies fall in the middle which justifies the location of mean, median, and mode scores. Your explanation also made me realize that a completely normal distribution must be fairly rare to organically come across in psychological measurement!

      Reply

    • Pawel Zawistowski
      Aug 28, 2020 @ 13:30:50

      Hi Cailee, I agree that both types of tests are important in their own way and serve a different purpose. I also agree that having a refresher on the normal distribution was very good to have. It’s been a long time since I took my stats class during my undergrad and I found it very beneficial to familiarize myself with it again and go over the terminology.

      Reply

    • Elias Pinto-Hernandez
      Aug 29, 2020 @ 16:13:28

      Hi Cailee,
      You summarized this week’s topic in a way that is clear to understand. However, I believe it would be a challenge for me at this time to explain any test results to a client. Nevertheless, we still have a way to get there. Great post!

      Reply

  12. Cassie Miller
    Aug 27, 2020 @ 13:46:40

    Criterion/domain referenced instruments focus more on an individual’s ability/performance when taking the given test/instrument. It allows for a representation of how well they understood the content/material. An example of this would be receiving a certain score for the answers correct on a chemistry exam, thus measuring the performers ability to accurately understand/interpret the information present. This type of instrument allows the performance of an individual to be compared to a pre-determined standard/criterion. There is also usually a mastery component for this instrument that allows for a pre-determined cut-off score that is used to identify if an individual has truly “mastered” that information. It determines an individual’s level of understanding and often identifies whether or not they are allowed to move forward in that given area/domain.

    Norm-referenced instruments differ in that they compare an individual’s performance to a “norm group” usually consisting of other individuals that share a specific commonality such as age, area of study, grade level, etc. Therefore, this instrument is comparing individuals performances with each other instead of to a specific criterion which occurs in criterion/domain referenced instruments. An example of this would be comparing Jane’s scores to Tommy’s scores on the same 5th grade Biology exam.

    It is very important to understand the difference between these two instruments because scores may vary significantly when using norm-referenced instruments, since individuals may receive a variety of scores. Thus, it is important to acknowledge that that person is only being compared to their peers which does not directly identify their own knowledge of the material. However, when using percentiles this instrument can be very important, as it provides more information about a given score and also acknowledges how much variation there is between peers. In addition, when using criterion/domain referenced instruments the focus on individuals performance in accordance with a given criterion allows for an understanding of how well prepared that individual is for that given domain. This becomes important when assessing someone’s qualifications in general. However, it is always helpful to compare both instruments together so that the tester can understand peer-relation in scores, as well as, individual progress.

    As far as normal distribution, I feel like I have a better grasp on it after watching your power-point. It makes sense that often times these distributions are skewed as it is very rare to have completely normalized data. It also makes sense that the larger the group the more likely the data will fit into a normal distribution. I am starting to gain a better understanding of z and t scores, but get a little confused when working with negative standard deviations and subtracting by 50 instead of 100. I’m assuming it is because it is in the first half of the normal distribution, but I am looking forward to further discussion in class on this topic. Perhaps more examples so we can attempt to figure out the scores and see if we are correct!

    Reply

    • Anna Lindgren
      Aug 29, 2020 @ 14:19:29

      Hi Cassie,
      Your definitions of criterion/domain and norm referenced are spot on and very in depth! I like how you point out as well that the two different types of instruments can yield different results even though the level of knowledge in that area might be the same, because of the comparison aspect of norm-referencing. You also rightfully point out that used together, you can get a more complete picture of a person’s understanding both individually and as part of a larger group.
      I’m with you on needing more practice with z scores and T scores, but I’m sure we will get plenty more before the midterm!

      Reply

    • Nicole Giannetto
      Aug 31, 2020 @ 14:38:50

      Hi Cassie! I liked how you included that there is often a mastery component for criterion/domain-referenced instruments that allows for a pre-determined cut-off score which is used to identify if an individual has truly “mastered” that information. I then enjoyed how you highlighted that understanding the differences between the two instruments is important because the individual taking a test needs to know the distinction in order to understand what their results specify or reflect on their own selves.

      Reply

  13. Christina DeMalia
    Aug 27, 2020 @ 14:53:55

    (1) Both norm-referenced and criterion/domain referenced instruments are used to measure something being assessed. The main difference between those two are how they assessed/what the score is compared to. In norm-referenced instruments, an individual’s score is compared to the score of a norming group, which is other individuals who took the same instrument in order to have a group to compare scores to. Criterion/domain instruments have scores with an establish standard. For example, an exam might require a score of 70 to be considered “passing”. The book also described this as a mastery level, and is a certain score that is expected, separate from anyone else who has taken the instrument.

    Looking at the difference between the two, my first thought was how the GRE is scored. When you get your results, you receive both a numerical score based on how many questions you answered correctly, and a percentile that shows how you compared with others who have taken the test. The GRE itself doesn’t determine a “mastery level”. However, many programs might state they expect a score that is over a certain cut off they’ve determined such as 150. In this case, it would be more similar to a criterion/domain instrument because there is a specific cutoff in score. Other programs might be looking for you to have scored above a certain percentile, such as 50th percentile. In this case, the scores are looked at in a more norm-referenced way because it is comparing your performance to the scores of other people who took the same exam. A more clear-cut example is that a test in school that is graded out of 100% would be criterion/domain.

    It is important to understand the difference between these two because how the results are assessed vary between them. Especially in psychological measurements, it is important that we understand how the scores are established so we can communicate to our clients what those scores mean. If an assessment is given and the result is a score of 48, we don’t automatically know what that suggests. If you were to tell someone that was their score, they might assume it is out of 100 like a grading scale (or a criterion-referenced instrument) and feel that they scored very low. However, if the score was actually from a norm-reference instrument, and they were in the 48th percentile, this would suggest they were right around the average score.

    (2) I feel that I have a pretty solid understanding of normal distributions. This topic was one I was able to learn about in my undergraduate statistics course, as well as revisit again in my Statistics for Social Sciences course. This recap refreshed and reinforced my understanding, so it is something I feel I have a solid grasp on.

    Reply

    • Christina DeMalia
      Aug 27, 2020 @ 16:24:53

      To add to number 2 of my response: My understanding of a normal distribution is that the mean, median, and mode are all the same. These results form a normal curve that is bell shaped and equal on both sides. A majority of the scores fall within one standard deviation above and below the mean, and only a very small amount of scores will be at the tail ends of the curve, 3 standard deviations above or below the mean.

      Reply

    • Alexa Berry
      Aug 29, 2020 @ 20:11:44

      Hi Christina,

      I like how you used the GRE as an example when explaining the difference between criterion and norm-referenced instruments. I never had to take the GRE so it was interesting to hear about how it is scored and how it relates to the material we are learning now. It’s unfortunate that some programs would set a mastery level when reviewing GRE scores, especially for students who don’t test well.

      Reply

  14. Brianna Walls
    Aug 27, 2020 @ 15:09:22

    1. A criterion-referenced instrument is used to measure whether or not someone meets the standard or criterion of a certain domain. The objective is not how the individual’s performance compares with another individuals performance but how they performed compared to the established standard or criterion that was established for that particular instrument. In other words a criterion/domain-referenced instrument measures if the individual has mastered the content or not (pass/fail). On the other hand norm-referenced instruments are used to compare an individual’s performance to groups of other individual’s performances who have taken the same instrument but it does not indicate whether or not the individual has mastered the content or not. It is important to understand the differences between the two instruments in order for someone to interpret the individual’s scores and what they actually mean in terms of how well the individuals know the content that is being measured or how well they did compared to other individuals being measured with the same instrument.
    2. In terms of normal distribution I understand the main concept of it. In a normal distribution of data there will be a bell shaped curve and it will be symmetrical. You can tell whether or not the distribution of scores is normalized because the mean, median and mode will all be the same score. This normal curve represents that the largest number of scores fall in the center and the rest of the scores decrease gradually in both directions. I also realize that the larger the group the more likely the scores will resemble a normal curve or the bell shape curve. All in all I have a good understanding of this concept but when you go over it during our zoom class meeting I’m sure I will have mastered the concept!

    Reply

    • Elizabeth Baker
      Aug 27, 2020 @ 16:03:17

      Hello Brianna,

      I really liked your explanation of the bell curve. I also liked your explanation of the norm-referenced instruments. It’s important to add that these instruments do not depict whether the test-taker has MASTERED the material. Even though, I took the SAT’s and ACT’s and felt that it was a test to see how well I’ve mastered the material. That’s just how I feel about any exam, regardless if it’s pass/fail or not. Great post!

      Reply

    • Anne Marie Lemieux
      Aug 29, 2020 @ 17:22:40

      Thank you for your explanation of the norm curve. The more I am reading everyone’s comments the more the normal curve becomes clear to me. I think I was overthinking it so I appreciate your simplification!

      Reply

    • Alexa Berry
      Aug 29, 2020 @ 20:07:43

      Hi Brianna, I liked your explanation of the bell curve and how we can use the visual to see if the data is normally distributed. When I was reviewing the bell curve information I also thought about how the norming group can effect the distribution of a bell curve. Like you said, larger groups make the bell curve appear to be a normal curve- this made me wonder what other ways a norming group can change the graph.

      Reply

  15. Zoe DiPinto
    Aug 27, 2020 @ 15:32:00

    A norm referenced instrument is a psychological test in which the raw score is compared to other test taker’s raw scores in order to be understood as a value. For example, new high school students who are applying to college take the SAT each year. The SAT has a measurement from 0-1600, however the student’s score only has value compared to the other student’s scores who are applying to the same college. Say you have a score of 1200. If the average Assumption student had an SAT score of 1400, that would make the value of your score decrease. However, if the average Assumption student had an SAT score of 800, the value of your score would increase.
    This is different from a criterion-referenced instrument test, which uses a standard measurement ranking to assign value to an individuals score. For example, its almost nationally understood that getting 95% of answers correct on an exam is an A while 75% correct is a C. The difference between these two testing styles is very important when we think about psychological measurement. There are some tests in this field where we are trying to compare individuals in a population to see if they need extra help, treatment, or challenges (especially in the education system). Other times, we use tests to certify people in trainings, award a degree or mastery component in which it doesn’t matter who the individual is being compared to.
    A normal distribution shows a sample of test scores that swell in the middle to show the majority and average scores fall in the middle. This makes sense to me as I have studied bell curves in undergraduate classes. I hope this class’s test grades fall into a negatively skewed distribution!

    Reply

    • Elizabeth Baker
      Aug 27, 2020 @ 15:58:28

      Hello Zoe,

      Your explanation of norm-referenced and criterion-referenced instruments was very clear! More clear than mine haha. I liked your explanation of these instruments, since you used real-life examples. At the end of your criterion-referenced instrument explanation, you added that the scores can depict which student/test-taker needs extra help. I thought that was a nice touch because it’s important for instructors to look over these assessment scores and help those who may have failed, so they can have a better grasp on the material and pass the exam.

      Reply

    • Cailee Norton
      Aug 28, 2020 @ 14:53:33

      Zoe,
      I really appreciate your application of the material to that of SAT scoring. For me, multiple examples to real world applications is how I understand material. I felt pretty good about my explanation of criterion-referenced instruments and norm-referenced instruments, however after reading yours I feel much more confident in that understanding! I think understanding what exactly the instruments are measuring is key in understanding what tests should be given and how those tests can be explained to our clients. Great job! I also appreciate your statistical humor in regards to our grades, I really hope you’re right lol!!

      Reply

  16. Elizabeth Baker
    Aug 27, 2020 @ 15:47:25

    I think I have a good grasp of the differences between these two types of instruments.
    Criterion/Domain Referenced Instruments, from what I understand, are assessments that are based on what someone should know by that certain age/grade. These assessments are to understand how well the test-taker can perform according to information they’ve learned in past/current courses/practices. For example, school exams would be considered as a criterion-referenced instrument because students are tested on what knowledge they have, and are continually tested to see if they’re grasping the information throughout the year/semester. These types of assessments can also be used to determine if an individual passes the class. Depending on the test-takers grade, the test-taker is either granted completion of the class or is told that they have failed and have to retake the class or exam. An example of this would be a test to get licensed in your field, or an entry exam.

    Norm-referenced instruments are assessments that are used to determine how much information the test-taker understands, but the results are used to show where the test-taker stands amongst other individuals who have taken the same assessment. I think an example of this would be MCAS, ACTs, and SATs. All of these assessments are there to show you how well you performed on the exam, and where you stand amongst others who have taken the same exam. Based on their performance, test-takers are either placed in the below average, average, or above-average percentile/section. These tests are also created to show how well you understand the information, but isn’t a pass/fail type of assessment.

    It’s important to know the difference of these two instruments because one is there to measure the test-takers ability to answer questions that their age/grade range should be able to answer (Criterion/Domain referenced instruments); and the other is there to give an idea of where test-takers stand amongst those who have taken the same exam, on a poor-to-above average scale (Norm-Referenced instruments).

    I think I understand the normal distribution curve, but using the curve to interpret scores is what confuses me the most. I understand that the curve is consisted of the mean, median, and mode, and can be used to interpret/explain how much the score(s) deviate from the mean. I understand how to calculate scores, but using the curve to further explain the scores is something I still struggle with.

    Reply

    • Christina DeMalia
      Aug 27, 2020 @ 16:39:54

      Standardized testing was also what popped into my mind when I read about norm-referenced instruments. I think the SATs, ACTs, etc. are a really good example of this because most, if not all, of us have had the chance to take an exam like this, and seen the percentile we fall into.

      I also think that criterion/domain referenced instruments are often used for measuring and comparing within age/grade ranges. I believe the textbook uses the word domain to describe this, which I think is a good fit because it also includes instruments that aren’t specific to just age or grade level. Your example of a licensing test falls into this, where the test might not compare you to a certain age, but rather a certain skill set that you are expected to have at that professional level,

      Reply

    • Cassie Miller
      Aug 27, 2020 @ 17:51:34

      Hi Elizabeth! I agree that it is important for a test taker to know where they stand amongst their peers (some could argue almost as much as whether they have met that criteria for the given domain). There were a few classes that I had taken over the years where the teacher struggled to communicate information effectively to the class and the distribution was skewed instead of being a normal curve. As a result, even though all of the students individuals scores reflected poorly when looking at the criterion reference instruments, it became obvious by comparing scores that all of the students were struggling. Thus, by using this norm-referenced instrument the teacher/professor can use this information to (hopefully) give the students a grading curve and begin presenting the material in a new way.

      I am also looking forward to learning more about using the curve to further explain scores past just the standard deviation because I struggle with this as well!

      Reply

    • Karlena Henry
      Aug 31, 2020 @ 15:12:22

      Hi Elizabeth,
      I was thinking about your example of the SAT for norm-based measurements, and I had a question. When you take the SAT, are your scores measured against other students, or is it based on your skill in preparation? When I think of a test that’s norm-based, I think of essay style exams, where the grader interprets the students understanding of the material.
      It has been a long time since I took the SAT, but that’s what I recall.

      Thank you!

      Reply

  17. Karlena Henry
    Aug 27, 2020 @ 16:56:08

    Criterion-referenced instruments are built to measure an individual’s skill in a specific way not taking other participants scores into consideration. For instance, a 16 year old wants to get their driver’s license. They prepare, and when they are ready will apply for a road test. The examiner has a series of tasks the applicant must complete in order to get their license. How the previous applicant did does not affect their performance. They need to display a level of mastery in order to be trusted to drive unobserved. If the applicant demonstrates the skill necessary, they will receive their license. If not, they won’t receive it. There is a clear dividing line on mastery. There are examples, however, that use criterion-referenced instruments without a clear goal. For instance, when a new student starts at their undergraduate program, the admissions office needs to determine their level of aptitude in order to place them in the correct class. This is especially important for returning students (like me!). Each level has certain skills that need to be demonstrated and the student is placed where they demonstrate their competence.

    Norm referenced measurements take into consideration of other people’s results when evaluating someone’s performance. In high school I auditioned for Western District Festival Chorus all four years. Each year I would prepare the same way, and then go in and show my skill. You would assume if I qualified for the chorus my freshman year, I would continue through the following year, but that was not the case. The judges would take into consideration all the applicants when designing the group. My Junior year I not only got accepted into All-District, but also All-State. The following year, I qualified for All-District but did not even get an audition for All-State. I didn’t perform any worse my senior year, but the other applicants demonstrated higher skill than the year before.

    I think normal distribution is important for gauging a person’s level of mastery for a given subject, but it shouldn’t be the only criteria. In the placement example above, knowing a student excels in the subject gives the examiner the opportunity to probe more, and find the best class to fit the student’s aptitude. I hope I’m getting this right!

    Reply

    • Christina DeMalia
      Aug 27, 2020 @ 20:54:33

      Hi Karlena,

      I really love the example of a road test for getting your driver’s license as a criterion/domain instrument. It is a perfect example for a time when a predetermined set of skills is being assessed. It also works well as an example because it is something many people have themselves gone through.

      I laughed a little at the thought of your performance on a road test being compared to how the previous applicant did. “Well they ran 4 stop signs but you only ran one so here’s your license.”

      I also like the example of auditioning for a norm-referenced instrument. It is a good point that skill level alone isn’t what’s being assessed, but instead one person’s skill compared to the rest of the group. It was very hard for me to get out of the mind frame of written formal exams when thinking of these so I appreciate your use of relatable examples that go beyond just written assessments.

      Reply

      • Karlena Henry
        Aug 31, 2020 @ 15:03:32

        Thank you Christina for your comments. When I was trying to find an example, I was looking for something people could relate to, and it seems I hit the mark! I think there are many tests that combine both criterion-based and norm-based. Especially in the field of psychology. I look forward to learning more.

        Reply

    • Cailee Norton
      Aug 28, 2020 @ 14:57:42

      Karlena,
      Like many others I really appreciate your use of the drivers license to explain that the mastery of driving is what matters for that criterion-referenced instruments. How they did in comparison to other drivers matters very little. In understanding this concept we are better equipped as helpers to provide answers to our clients when discussing the results from their assessments. I think you also raise a great point about normal distribution. While a student may fall in a particular percentile, it is important to have an understanding of our client to gauge how this can impact further treatment or paths to be taken. Thank you for your insightful examples and connecting it to outside of just material, but to our basic goal for this program: helping others!

      Reply

  18. Anne Marie Lemieux
    Aug 27, 2020 @ 17:07:00

    My understanding of the difference between norm referenced instruments and criterion/domain referenced instruments is significant. Norm-referenced instruments compare one’s score to others who have also taken the assessment. It is a way to assess how one compares to the norm group. For example, the Woodcock-Johnson compares the testee to other same aged peers performances. While criterion-referenced instruments compare one’s score against a set measuring tool. That measuring tool is predetermined. For example, the kindergarten screening establishes whether an incoming student has mastered letter identification. However, this becomes more difficult to measure as the task that is being assessed becomes more complex. For example, my friend missed passing her realtor’s license by one point. However, she may have just as good of an understanding of real estate as a peer who scored one point higher than her.

    My thoughts on understanding norm distribution are that I think I understand it overall. However, I often get overwhelmed when statistics start being discussed, so I could be way off. When I think of normal distributions I think of the baby weight and height chart. It is measuring something definitive. If a baby is in the 50% percentile for weight and 90% for height, it’s easy to understand that they are of average weight. It is also simple to establish that the child is above the average in height. However, I get confused when using 50% as the mean verses 68%. I would assume that the baby in the 90% percentile is two standard deviations above the median for height but I’m not certain. I am clear on understanding that if a baby was two standard deviations below the mean for weight may have failure to thrive or be a premie. I am also unsure if you use the mean or median when discussing this. I know the definitions for each but have difficulty when putting them into practice. I look forward to gaining more clarification in class.

    Reply

    • Carly Moris
      Aug 29, 2020 @ 15:21:33

      Hi Anne Marie! You seem to have a really good understanding of norm and criterion referenced instruments. I like how you included example tests in your explanation. I also liked the example you gave about your friend not passing her realtor license by one point. This shows how the cut off for some criterion referenced tests can be a bit arbitrary, because does one point really make a difference between mastering a subject and not?

      I also agree with you that statistics can get a bit overwhelming. I’m not sure if child weight and height would be considered a normal distribution. But when looking at a normal distribution it is important to remember that the mean, median, and mode are the same. So you can use both mean or median when discussing a normal distribution, because they should be referencing the same point in the data. This means that half of your data should fall above and below your median and mean. That is why in a normal distribution you use 50% as the mean instead of 68%, because the mean should mark the middle of the distribution. Something 2 standards deviations above the mean would also be considered to be in the 98th percentile. I hope this helped make things clearer!

      Reply

  19. Anna Lindgren
    Aug 27, 2020 @ 17:32:48

    A criterion or domain-referenced instrument is an assessment that tests a person’s knowledge in a particular area. The assessment is made by experts in that field or area, and they may determine a cutoff score for whether or not a person has mastered that subject, as in the case of a licensure exam. A norm-referenced instrument is an assessment in which a person’s individual score is compared to a group of people who have taken the same instrument. It is important to understand the difference between the two because criterion or domain-referenced only has to do with one individual’s score and perhaps whether or not they “passed” the cutoff point, whereas a norm-referenced instrument tells us where that person’s score fits in with the norming group. It’s my understanding that in our clinical work, we will be using more norm-referenced instruments of assessment than criterion-referenced, except perhaps in the case of school counselors who may do a lot of criterion-referenced assessments with their students.

    I have a pretty good handle on the concept of the normal distribution, and how in those cases the mean, median and mode are at the same central point of the data and individual scores will lie somewhere on that symmetrical bell curve, within a certain amount of the standard deviation. I think once I practice using it with some data sets I’ll have a better grasp on it (it’s been a while since I took stats!).

    Reply

  20. Alexa Berry
    Aug 27, 2020 @ 17:33:12

    Norm-referenced instruments compare the score of an individual with the scores of other individuals. Norm referenced instruments, all that are being assessed have taken the same instrument. Examples of a norm-referenced test include tests that are graded on a curve. Criterion-referenced instruments compare an individual’s score with a prior established standard, such as in standardized tests like driving tests or AP exams. It is important to understand the difference between these two types of instruments because the way these instruments are implemented can have an effect on the results. For example, since norm-referenced instruments compare the individual to the group, one person’s answer may be better than the rest of the answers provided by the group and can be ranked as such. In contrast, in criterion-referenced instruments everyone can provide a correct answer and it is not dependent upon other responses. Relating back to the driving test example, this is why during driving tests individuals are assessed based on their ability to drive, not their ability to drive better than other people. If driving tests were norm-referenced this would affect the amount of people who are able to pass the test.
    Regarding normal distribution, I am comfortable with my current understanding. Something I did think of when reading the text is how the norming group can affect the normal distribution. As stated in the text, there is no universal standards for what constitutes a good norming group. Thus, being familiar with the characteristics of a population in a norming group is a responsibility that falls on the practitioner using an instrument.

    Reply

    • Connor Belland
      Aug 28, 2020 @ 19:49:42

      Hi Alexa, Great response! I like your example of the driving test as a criterion referenced instrument. I think it is very important that driving test scores are compared to an established standard because imagine if drivers were just compared to how well they do compared to other test takers. What if all the test takers that week were terrible drivers but they did better than last weeks really terrible drivers. We would have a lot of bad drivers on the road, Which is why it is so important to choose the right type of reference instrument depending on the test.

      Reply

  21. Nicole Giannetto
    Aug 27, 2020 @ 17:48:27

    1. Both criterion/domain-referenced and norm-referenced instruments are very similar to each other, but have some major differences that set them apart. In criterion/domain-referenced instruments, the individual’s score gets compared to an established standard. For this type of instrument, the goal is not to focus on how well the individual did on the test compared to others, but on how the individual scores in regards to the standard. One example of this type of test could measure the performance of a student against standard learning measures that reflect the stage the student is in within their educational career. This way of measuring and scoring performance is different for norm-referenced instruments. In a norm-referenced instrument the score of the individual taking the test is compared against scores of other individuals who took the same instrument, and not against an established standard. Norming groups can be much larger or much smaller depending on the reference sample size that is used. Examples of a norm-referenced test are the SAT and ACT, because the score of each student is compared to an average based on samples that identified a norm group. I think it is important to know the differences between these two instruments because on the one hand, criterion/domain-referenced instruments tell us if a student met or passed a standard, while norm-referenced instruments are able to tell us how an individual performed compared to average scores of other individuals who took the same instrument. What both of these tests show is that measuring constructs can create different results depending on what researchers want to understand regarding the behavior they are testing.

    2. I am familiar with and fairly comfortable with my understanding of normal distribution. Generally, it does make sense for me. One aspect that I wish to work on is being able to memorize the percentages and numbers that describe various sections of the curve and its standard deviations.

    Reply

    • Cassie Miller
      Aug 27, 2020 @ 18:10:04

      Hi Nicole! I like the idea that you brought up in your response about how these measuring constructs can be strategically used based on the type of information an individual is looking for. I think it is very interesting to think about using a specific construct of measurement for different aspects of behavior. For example, someone might be interested in measuring hand raising behavior for a 6th grade class against the hand raising behavior of a ninth grade class. Here the researcher wants to compare the data from both peer groups which would mean he is using a norm reference instrument. However, if the teacher was tallying each time a child raised his/her hand and had set a previous standard of participating in this behavior at least three times during class to earn participation credit, she could use a criterion/domain referenced instrument to assess whether each individual student is meeting that given criteria.

      Reply

    • Anne Marie Lemieux
      Aug 27, 2020 @ 18:45:32

      Your explanation really simplifies the differences between the testing formats. I agree that criterion/domain-referenced instruments may be a better way to assess individual progress. It can allow you to set a baseline for the testee and establish if they are making effective progress or regressing in the area that they are being measured. While norm-referenced instruments compare the tester to the norm group only telling you how they compare to others. The could be below the average but still be making effective progress. Thanks for your insight.

      Reply

    • Connor Belland
      Aug 28, 2020 @ 19:42:52

      Hi Nicole, Your examples of the two different types of instruments really helped me to visualize a real world application of each one. I think its important that you mentioned the varying sample sizes in norm-referenced tests can really affect it. That is probably a big reason that norm referenced instruments should only be used in certain situations depending on the test. I agree with you that both instruments are very useful in school settings to test not only the individuals performance but the schools performance compared to others.

      Reply

    • Brianna Walls
      Aug 29, 2020 @ 15:49:35

      Hi Nicole! I liked how you explained in your own words how both of the instruments are measuring constructs but can create different results depending on what the researchers want to understand regarding the behavior they are testing. I also hope to become more familiar with the percentages and standard deviations regarding normal distribution!

      Reply

  22. Timothy Cody
    Aug 27, 2020 @ 18:42:40

    1. A Criterion/domain-referenced instrument tests people based on what they know and does not compare their scores to others. It will be based in boundaries that section an individual’s score and will put them in a percentile in which they have an idea of how well they performed. This instrument approach allows individuals to excel to a mastery of their test performance when they exceed a certain cutoff score. A norm-referenced instrument, however, compares scores to other individuals who have taken the same test. It is important to know the difference between these two instruments because some tests when going for a license in Counseling use a domain-referenced instrument, which will determine your mastery of the exam, but will not compare the results to other test takers. This allows no bias between two individuals taking the same exam.
    2. From what I understand, a normal distribution, also known as normal curve, is a curve on a graph where you can compare your percentile rank in comparison to the mean. It makes sense because you can compare your score to what would be the standard or average score, and where you would fall on the percentile. It makes sense to me as to how the scores would fall on the normal curve based on their standard deviation. It makes even more sense when scores are positively or negatively skewed, for the mean is skewed based on if there is extreme data in one direction or another.

    Reply

    • Carly Moris
      Aug 29, 2020 @ 14:44:05

      Hi Tim! I think you have a good explanation for the difference between norm and criterion referenced instruments. You bring up an interesting point that criterion referenced tests can prevent a bias between test takers because their success will be based on a predetermined cutoff point, not on how other individuals have scored. But I also think it is important to understand the difference between norm and criterion referenced instruments, because the type of instrument will effect how you interpret the results.

      Reply

  23. Connor Belland
    Aug 27, 2020 @ 19:41:16

    1. The main difference between Norm-referenced instruments and Criterion/domain instruments is that in a norm referenced instrument, the score is compared with other people that have taken the same test. Often times scores are compared to an average to see how they match-up against other individuals. Criterion scores on the other hand are compared with an already decided upon standard. I think it is important to know the difference between these two instruments because it is important to pick the right one to use in a given situation to get the most accurate comparison. I think both are important in school testing because it is important to see students scores compared to both the criterion/standard and also to see how they did compared to their classmates. Especially when comparing scores on standardized tests it is important to see how students did individually compared to the set criterion and also to see how their score compared to the rest of the students school.

    2. I think i have a good general understanding of normal distributions and curves. I definitely need some refreshing on the subject but it should all come right back to me. I have a decent understanding of where standard deviations and some percentiles fall on the bell curve but i definitely want to try to memorize more of them though. I would also like to get a better understanding of how I will use normal distributions in real life situations and how to interpret and understand real world information and scores, especially in the school setting.

    Reply

    • Tanya Nair
      Aug 28, 2020 @ 19:49:35

      Thank you for your comment, Connor! I think you did a great job summarizing the two different measurements. I think it is important how you want to get a better understanding of the normal distribution and how it can be applicable to real-life settings. Often times when I learn things, I also want to know how it can be applicable to real-life situations because it helps me understand the material better and allows me to generalize the knowledge within various settings. It is interesting that you want to learn specifically for school settings. What is it that makes you specifically interested in only this, if you don’t mind me asking?

      Reply

    • Timothy Cody
      Aug 28, 2020 @ 21:05:57

      Hi Connor! My suggestion for understanding the normal distribution better is use it in real life situations, just like we were doing with intelligence tests. For school settings perhaps using them with a classrooms exam scores. you can find the mean for a specific exam and use a z test in order to determine one student’s percentile compared to the mean. Be careful if there is a positive or negative skew with your distribution curve, though. You don’t want to skew your data in one direction or another.

      Reply

  24. Maya Lopez
    Aug 29, 2020 @ 20:44:33

    I know I am a bit late on this post but this is NOT a pattern of mine and I am sorry. (1) Criterion/domain-referenced instruments to my understanding are used by comparing a score you get to what is expected to know. Say, one took a science exam and scored a 72% but the passing grade is 75%, everyone taking the exam is expected to know 75% correct on the exam and will all be compared to 75%. Norm-referenced instruments are used by comparing results to the norming group which as we know is the group that was used to develop the tool/ test. An example of this would be the ACTs where our scores are compared to others who are testing. It can be useful to know the difference because when receiving a score whether for oneself or for another, we should be able to determine what to compare that score to so it is not just an arbitrary number. (2) My thoughts on understanding the normal distribution are fine, I don’t feel too confident but I do understand the skew of positive or negative is determined by which way the tail is facing, I know I will have a lot to learn from this course.

    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 68 other followers

%d bloggers like this: