Topic 2: Reliability {by 9/16}

Based on the text readings and lecture recording due this week consider the following two discussion points: (1) Correlation does not equal causation! Share your thoughts on why this assumption is still a common mistake, especially in the mental health field.  (2) Share your thoughts on why reliability is so (wicked) important for psychological assessments (this is a bit of a deep question – give it your best shot).

 

Your original post should be posted by 9/16.  Post your two replies no later than 9/18.  *Please remember to click the “reply” button when posting a reply.  This makes it easier for the reader to follow the blog postings.

47 Comments (+add yours?)

  1. Mary Altomare
    Sep 14, 2021 @ 08:44:33

    After listening to the lecture, reading the textbook and my own understanding of correlation, I feel as though this assumption is still a common mistake, specifically in the mental health field because individuals want clear cut answers. Although the two variables may have a significant relationship with one another, there are other variables that also play a significant role that is affecting the relationship. I think Dr. V gave very good examples of this, particularly with exposure to TV violence and aggression. These two variables do make complete sense and I believe in people’s minds, it’s easier to digest that if certain individuals reduce their expose to TV violence, their aggression will go down. However, as Dr. V explained about correlations, we cannot determine if exposure to TV violence is causing aggressive behavior or if aggressive behavior is causing exposure to TV violence. Furthermore, as coefficient of determination explains, these two variables represent a small portion, and that there are several unknown variables that are not being examined, that have a significant effect on the relationship.

    When I first started as a therapeutic mentor, I would hear from parents that their kiddo was engaging in aggressive behavior due to playing violent video games. This made complete sense to me, and I figured if I assisted the kiddo with finding different activities to engage in I would have helped decrease their aggressive behavior. Boy was I wrong! The more time I spent with the kiddo, I realized the kiddos would mirror their aggressive behaviors of their parents, such as the way they spoke about one another and how they handled disagreements. Also, I observed kiddos become escalated when their parents would update me in front of the kiddo, on a recent incident that occurred. I quickly picked up on the fact that kiddo became anxious when their parents would talk about their behaviors in front of them, but instead of speaking on their feelings, they would engage in aggressive behavior. Moral of the story, I have realized that there are SO many variables that affect an individual’s behaviors or mental health and it is imperative to look at the whole picture not just a small portion. I just happen to work with a lot of kiddos that present with aggressive behavior and so this example Dr. V gave, truly resonated with me.

    I believe reliability is so important is because in the field of psychology we rely heavily on assessment (at least that is what I believe.) Therefore, as future clinicians it is imperative for our treatment that the score our clients are receiving on the assessments we are utilizing is dependable. As Dr. V explains, we can acknowledge that the assessments are not perfect and that our client’s scores do include some error. With that being said, it is critical that the assessments are close to reliable as possible, because we can then infer that the scores are going to be close to consistent. This is important for clinicians because we want to make sure that our clients are being assessed properly in order to provide quality and effective treatment.

    Reply

    • Kristin Blair
      Sep 14, 2021 @ 10:05:09

      Hi Mary,

      I absolutely loved your explanation of the violent behavior due to video games. I love that you could draw from your own personal experience to explain your point. My ABA kiddo was showing similar behaviors and the mother would discuss his poor behaviors to me in front of him and I too, could see him becoming anxious and sometimes angry because of this. I struggled with wanting to tell the mother this because, well, you never want to tell someone how to parent their child but, I could see that it was it was causing more harm than good and didn’t really find it necessary. I spoke with my kiddo privately about the issue and he expressed to me many things that extended way beyond the simple fact of “his video games make him aggressive”. It seemed more about the fact that he was never allowed to play the games and felt as though he did not have access to any preferred activities which would make anyone a little frustrated! So I very much agree with you saying that there are SO many variables that affect and individuals behavior and mental health.

      Reply

    • Vanessa Nichols
      Sep 14, 2021 @ 12:30:02

      Hi Mary,
      I think your response was great, especially your personal example of working with those children. I agree with you. For people who do not deal with variables and the scientific method, it is just easier to think that Variable A caused Variable B, so get rid of Variable A and Variable B will also change. But based on your personal example, we know that behavior is way more complicated than that. There can be seven other variables you weren’t even aware of that are all affecting variable B.

      I think your reliability answer is excellent. It’s also important to remember that without reliability (specifically re-test for this example), the assessment may not even measure what it is supposed to be measuring. Reliability is the backbone of assessment, and assessment helps clinicians make important decisions and conclusions regarding the client. If the assessment is not reliable, then neither are those decisions or conclusions.
      Thank you!

      Reply

    • Victoria Cestodio
      Sep 14, 2021 @ 13:20:36

      Hi Mary,
      I loved how you talked about your experience as a therapeutic mentor! I found the story you told super interesting. It is also a great example of correlation not equaling causation. Your mentee behaving aggressively because of the video games he was playing would make sense to me too, but also I would think it has to do something with the people he is around which was also the case. This example shows that there is so much more going on than one small thing causing another.
      I have a similar experience to you. I used to work in an elementary school in a first grade classroom. One student was constantly very aggressive with teachers and classmates and his parents would always tell the teacher it was because of his ADHD, which made complete sense to us. However we later found out that his parents were having marital problems and were getting divorced which could be another cause for his aggression. I feel like this is extremely important for the field we are going into!

      Thanks for sharing your own personal experience!

      Reply

    • Bekah Riley
      Sep 14, 2021 @ 19:18:56

      Hi Mary,

      I thought your post this week was very informational! I liked how you started off by stating why the assumption that correlation and causation are equal is commonly made in the mental health field because people want clear cut answers. I had the same idea! I also really liked how you explained the example that Dr. V gave! I thought this example really clearly depicted how different things can correlate, but that does not mean they are the cause of each other.

      I also really appreciated the inclusion of the experience you had as a therapeutic mentor in your post! This example really showed how two things may correlate together, but an outside factor could actually be the cause. In this case, the violent video games did not cause violent behavior in kids, the mirroring of their parents behavior was the cause.

      Reply

    • Tom Mandozzi
      Sep 15, 2021 @ 20:35:34

      Hi Mary,

      I really resonated with your example from your work as a Therapeutic Mentor. I also work as a TM and this is absolutely true; I have seen the same scenario! I think when dealing with aggressive behaviors, having an outside and comprehensive picture of what is going on is so important. When dealing with a child’s behavior, I have seen time and time again that the parents often attribute the aggression to media such as tv shows or videogames that kids are watching, but rarely look at themselves for their role in the broader family dynamic. The parents make sense of the behavior through the idea that violent videogames “cause” aggressive behaviors, but there are so many more variables at play here that are contributing to these behaviors even if there is a correlation between the variables. Great points here!

      Reply

  2. Kristin Blair
    Sep 14, 2021 @ 09:52:31

    Correlation does not equal causation is basically saying that, just because two sets of data appear to be positively correlated, it doesn’t always necessarily mean that one causes the other to occur. For example, yearly shark attacks and ice cream sales tend to spike at the same time. That is not to say that consuming more ice cream makes shark consume more people, or vice versa?! Therefore, we must be careful of this when conducting research and interpreting data and consider external factors that may make this true. Perhaps it could be that these both take place in summer months, which would make sense. The less information that we have about an assessment instrument, means we tend to rely on the basic correlations more. This is why it becomes very important to record as much information about the test subjects and environment as possible when conducting an assessment. More information helps us look at correlated data in a more informed manner so we can make the most accurate assessment of the information.

    Another example of this, which is sort of a popular topic right now, are vaccinations. There are many people who have stated that there is a link between vaccinations and Autism. Jenny McCarthy wrote a book about it, and so have several others. However, there really haven’t been many scientific findings to support that this is in fact true. Parents are naturally very concerned for the health and well-being of their children and therefor do not what to be immediately dismissive of a claim like this, so they may feed into a bit, which is then influenced by confirmation bias. I work as an ABA therapist and work with children who have Autism daily, and I know that the diagnosis of Autism is a complex process and not something that is a simple blood test. Children are typically formally diagnosed between 9-12, ages where they typically receive many vaccinations, but first signs can be seen as early as 12 months of age. The problem then lies in the actions people take when they interpret these findings inaccurately; such as, skipping or waiting on vaccinating their child. This then leads to a spike in things like whooping cough and the measles, which you can see in past statistical data.

    Reliability is the consistency of the results in research. Therefore, in order for research to be reliable, it should show the same (or similar) results if repeated. Reliability is so important for psychological assessments because ensures your testing is adequately contributing to your hypothesis. It also ensures that your results are due to the study and not any additional variables. Furthermore, if a study is reliable, it makes it a widely accepted and therefore used more consistently. A commonly used example for this that many people resonate with is a bathroom scale. If you stand on your scale every single day, you will assume it will give you a fairly similar reading. A scale would be entirely useless if it gave you a random number every day. However, this does not reflect validity. Reliability is mainly referring to the consistency of the result. You could have a scale that is not “calibrated” correctly and it shows your weight as being 15 pounds less than normal. If you stepped on that scale over and over again and it still showed the same number, it would still be considered reliable, but not accurate. Hence why reliability and validity go hand and hand and having both is ideal for the most accurate results.

    Reply

    • Emily Barefield
      Sep 16, 2021 @ 11:27:07

      Hi Kristin,

      You did a really great job of explaining how correlation and causation are not equivalent and why this matters. Your example about sharks and ice cream showed highlighted the absurdity of the conclusions that can be made based on this assumption, and your example with vaccines and Autism showed just how dangerous this reasoning can be. In some cases assuming correlation and causation are the same thing can lead to misinformed and costly decisions. I appreciate that you brought in your experience working as an ABA therapist in this discussion.

      You also did a good job of explaining reliability and providing examples. I appreciate you tying reliability to research using psychological assessments. Good post!

      -Emily

      Reply

  3. Emily Barefield
    Sep 14, 2021 @ 10:59:15

    Correlation is often mistakenly equated to causation in the mental health field because, in most cases, the causes of mental illnesses are unknown. We are aware of many factors that may contribute to the development of mental illnesses, but we have not identified a cause (or causes) for mental illnesses. Unlike many physical illnesses, where a pathogen, for example, can be identified as the cause of the illness, mental illnesses do not have a single identifiable cause. Despite this, the field adheres to a medical model of mental illnesses, meaning mental illnesses are thought of and conceptualized the same way as physical illnesses. Mental illnesses are described to those in the field and to the public the same way that physical illnesses are, and so when something is associated with an increase in frequency or severity of a mental illness it is thought of as a cause. Conflating correlation with causation in the mental health field occurs because of how we conceptualize and discuss mental illnesses.

    Reliability is important for psychological assessments because these assessments are used to make potentially life-changing decisions for those who take them. Many psychological assessments are used in the process of diagnosing individuals with mental illness. Receiving a diagnosis affects whether an individual receives treatment and often if that treatment is covered by insurance. It is important that an individual’s scores on an assessment are as close to their true scores as possible. This allows for the therapist to develop a treatment plan that is best suited for the client’s needs. A treatment planned developed from an unreliable assessment is likely not able to meet the needs of the client.

    Additionally, an assessment needs to be reliable before its validity can even be considered. It is impossible for an assessment that is unreliable and has inconsistent scores to be measuring what it was intended to measure well. It is important that an assessment is measuring the construct it was intended, so that the client is receiving treatment that best suits their needs.

    Reply

    • Vanessa Nichols
      Sep 14, 2021 @ 12:43:45

      Hi Emily,
      Wow, I think your answer, especially to the first part, is fantastic. I think you are entirely correct. Because physical health is so easy to understand for people, a lot of people think the mental health field works the same way. For example, in the physical health field, I broke my arm, so I have a cast, which will cause my bones to heal correctly. People who have mental illness are always looking for their cast, but the mental health field is not clear-cut. I completely agree with you that because people are used to dealing with physical illness in such a manner and because they conceptualize their mental illness to be like that, they want correlation to be as simple as causation, but that is not how it works.
      I also like your answer to the importance of reliability, and it’s very similar to mine. The mental health field uses these assessments to draw significant conclusions for clients. If the assessments are not measuring what they should be or are not consistent across the board, the information provided to this client could be completely wrong. This can lead to misdiagnosis, worsening of symptoms, and loss of the client-therapist relationship. Thank you!

      Reply

    • Victoria Cestodio
      Sep 14, 2021 @ 13:55:28

      Hi Emily,
      I totally agree with you when you say that assessments are so crucial and important because they make life changing decisions for the client. The reliability of the test is key because how the client then receives treatment is based on their scores. If the test isn’t reliable we are treating someone completely wrong which will then serve them no good, and won’t provide them positive results. Reliability is crucial for our field and I think talking about this more has made it even more apparent to me.

      Great job!
      Victoria

      Reply

    • Bekah Riley
      Sep 14, 2021 @ 19:06:53

      Hi Emily,

      I really liked your post this week! I thought your description on how correlation and causation are mistaken for one another in the mental health field was very clear and informational. When creating my post, I also reflected on how many mental illnesses do not have a specific cause, and there are so many different factors that can contribute to different mental disorders. I really liked how you tied the example of the medical model into your response and how in this model, mental and physical illness go hand in hand when determining a specific cause. This is a valid reason to why many people may confuse correlation with causation; it is just how mental illness is depicted!

      I also loved how you included that assessments administered by clinicians can be life changing for clients. This effects the client’s outcome in terms of diagnosis and treatment. That is why the reliability of the assessment is so important!

      Reply

    • Mary Altomare
      Sep 16, 2021 @ 07:59:21

      Hi Emily,
      I really liked your description on how correlation and causation are mistaken for one another in the mental health field. As we learn more in Dr. Doerfler’s class about the medical model and the way it is utilize to explain mental illness, it makes complete sense why people mistaken correlation and causation. In addition, I agree completely with you that reliability is imperative for psychological assessments because these assessments are utilized to assess people and create an effective treatment plan that is best suited for the client’s needs. However, if the assessment is not reliable that it can be detrimental to a person’s treatment.

      Reply

    • Tressa Novack
      Sep 16, 2021 @ 15:08:11

      Hi Emily,
      You point out something that I did not even think about when answering the question about correlations, which is that the mental health field still relies on the medical model. It would make sense that we would look for one singular cause for a mental illness like we do with physical illnesses since we rely on the medical model. I can see how that can lead to people, including mental health professionals, to confuse correlation with causation. This is a really great point, that completely went over my head before reading your post. I agree with your thoughts on reliability. I also pointed out how important it is for creating treatment plans. Something I added in my post is that reliability is also important for research. If we do not have reliability in experiments then researchers cannot draw conclusions from results or generalize them to any populations.
      Tressa

      Reply

  4. Vanessa Nichols
    Sep 14, 2021 @ 11:56:58

    Through the text and the lectures, I have learned the correlation/correlation coefficients are a tool of reliability to show the numerical indicator of the relationship between two variables. Something I already knew about correlation was that correlation does not equal causation. This means just because the two variables have a relationship( positive or negative correlation) does not mean that one variable caused the other to go up or down. I think there are multiple reasons that people still confuse correlation for causation. I believe that people who are not used to variables and the scientific method forget that other variables are still at play. So it is not as simple as A+ B = action because there can be a third variable they didn’t even know existed. For example, if A equal Weight loss and B equal TV time. We see that A and B are negatively correlated. Some people would assume that because tv time when down and weight loss went up, variable B has to be the cause. They don’t see that variable C playtime went up as well and could affect both variables.
    I think the second reason people confuse correlation and causation is because causation is easier to understand and wraps it all up finished with a nice bow. Correlation is more complicated. I think, especially regarding mental health, people want clear-cut answers. They want to know what is affecting, how to treat it, and, most importantly, why it affects them. In mental health, specifically, the why can get very confusing. So I think it is just easier for some people to believe that there is always a cause that can be seen.

    Reliability is so important because it is the backbone of the assessments that clinicians give. These assessments help clinicians make essential choices regarding diagnosis, treatment, and severity. Reliability is the consistency of these measurements so that they can be given repeatedly to groups without finding any significant differences. Reliability is vital for filtering out bias. Knowing the reliability of measures can help clinicians select which instrument to use (higher reliability usually means a better tool). Imagine, as a therapist; you are giving a depression inventory assessment to a client. If the assessment is low on reliability, first, you cant trust that your finding of your client being low or high on depression is correct. This error can lead to misdiagnosis. Second, you can not even trust that the assessment consistently measures depression or depression variables at all. So by using an assessment with low reliability, you risk going in the wrong direction regarding treatment and diagnosis. This misdirection can affect your credibility and client-therapist relationship.
    Reliability also helps when interpreting the results. Without reliability, scores like the norm mean and median mean nothing because they haven’t been consistent through different group testing. Without a consistent mean or median than we have nothing to compare the new scores to. This causes the scores to be ambiguous and impossible to draw conclusions from.

    Reply

    • Mary Altomare
      Sep 16, 2021 @ 08:25:07

      Hi Victoria,
      I thought your post was awesome this week! Your description of correlation and causation was informational. I agree with you that it is easier to understand causation versus correlation. I think as individuals it’s simpler to understand two variables versus multiple, especially when we want to figure out how to fix a certain issue. It may be overwhelming (and I can attest) when there are multiple variables involved to solve and figure out in order to resolve a problem. I completely agree with you that an assessment reliability is crucial for diagnosing and providing treatment for an individual. If the assessment is unreliable, we as clinicians are not reliable in our treatment to our clients!

      Reply

  5. Victoria Cestodio
    Sep 14, 2021 @ 13:08:24

    Within my time in undergrad my professors would always stress “correlation does not equal causation”. However, we still see this as a common assumption by many people but it is wrong! Just because two things tend to be positively correlated, does not mean that x causes y or y causes x, there could be a multitude of other factors. One correlation that I have seen is that ice cream sales and sunglasses sold are positively correlated. Just because someone is buying sunglasses does not mean they are more likely to get ice cream, or vice versa. A factor outside of X and Y could be the weather which could be variable Z. More people tend to get ice cream when the sun is shining, and the same thing goes for sunglasses.

    Dr. V gave great examples in the lecture such as the one with tv violence and aggression. Someone’s aggression could be leading them to watch violent tv or the violent tv could be making them have more aggression. However, variable Z can also come into play, which could be their parents are showing more aggression in the household, etc. One I also thought of was smoking leading to lung cancer. Smoking can definitely put you more at risk for lung cancer, but we cannot definitively say that smoking will/has caused it. Other things could cause it like pollution for example (variable Z).

    When it comes to the mental health field I feel like this is still such a common assumption because we want an answer that is clear and tells us exactly what is going on with an individual and/or a group. Therefore, we tend to forget or not want to dive deeper into other possible variables that could be causing an outcome. The clear answer to this is an outcome is never being caused by one specific thing, there are a multitude of reasons and variables.

    Reliability is super important for psychological assessments because if our psychological tests do not have reliability it wouldn’t be beneficial for us because it would have constant inconsistency. If the test is inconsistent we can’t rely on it, making results harder to get. If our tests yield different results each time, it is not reliable and then we also don’t have validity (we can’t have validity without reliability first). Therefore, having a reliable test is the first step to having good outcomes for researchers. As future clinicians if the test we use is not reliable we will not know how to effectively treat our clients which can be detrimental.

    Reply

    • Emily Barefield
      Sep 16, 2021 @ 11:40:55

      Hi Victoria,

      I’m glad you highlighted that “correlation does not equal causation” is a concept that a lot of people are familiar with or have heard of before, but seem to forget in many situations. I appreciate all the examples you gave, they are really helpful for showing that there is often additional factors that affect the relationship between two variables. I think you are absolutely correct in saying that people want simple answers regarding mental health and so look for straightforward answers even if those type of answers are not accurate.

      You also did a good job of explaining reliability and bringing up that we cannot have a valid test without having a reliable test. Additionally, we need a reliable (and valid) test to adequately diagnose and treat clients. Great post!

      -Emily

      Reply

    • Monika Dhamale
      Sep 17, 2021 @ 18:27:24

      Hello Victoria,
      The explanation part for correlation does not equal causation is on point. The many examples you provided explain how there can be multiple other factors affecting a particular variable which we might overlook or don’t take into consideration before making assumptions about the cause and effect. I totally agree with you that when it comes to mental health field we want clear answers and sometimes we tend to jump to conclusions in seeking these answers. Great job explaining the importance of reliability too.

      Reply

  6. Bekah Riley
    Sep 14, 2021 @ 18:55:08

    Correlation and causation are not equal to one another, but the assumption that they are is very commonly made, especially in the mental health field. To understand why correlation and causation are not equal, it is important to first understand what each of them are independently. After reading the chapter and listening to the lecture recordings, I have an understanding that correlation examines different scores and the relationship between them, whether that relationship is positive or negative. Causation on the other hand shows that one score is the result of another score. Correlation is often associated with causation because if scores positively correlate, for example, it is assumed that one score leads to another because they are both increasing together. A specific example of this may be a rise in textbook sales at Assumption University (AU) and a rise in female students attending AU occur within the same timeframe. Although this is a positive correlation, it does not necessarily mean that the rise in book sales went up because there are more female students attending the University and vice versa.

    In the mental health field, correlation and causation are especially mistaken to be equal to one another. This is because it is natural to want a clear cause to a diagnosis, however, often times in mental health disorders there is not a clear cause. So, if studies on mental health disorders show a positive or negative correlation, it does not necessarily mean that one factor in a study is the cause of another because there could be many other factors that change the outcome of the study.

    Reliability is so, or in other words “wicked” important for psychological assessments for a number of reasons. In order for clinicians to administer assessments, it is important that the assessment is reliable/consistent for purposes such as diagnosing/treating a client effectively to ensure they have the best possible outcome. In addition, it is important to consider the evaluation of reliability coefficients. There are not certain assessments/approaches that are generally preferred when estimating reliability. This is because the reliability is very dependent on the individual or population, which then leads to the type of assessment and the purpose of the assessment for a particular client.

    Reply

    • Lauren Pereira
      Sep 15, 2021 @ 14:16:11

      Bekah,
      Your description between correlation and causation is very thorough and brings up a lot of good points. You did a great job describing both of these terms separately so that it is easier to determine their differences. I also find your example to be helpful and a good way of showing positive correlation where the two variables do not rely on one another. They can both have the same outcomes but this is for different reasons which is important to realize in psychology.

      I also like how you talked about reliability. It is so important to recognize because it is what is needed to get a more accurate and dependable result. Without this, we may not be seeing such significant changes or increases within variables or clients. Being able to clearly see these reasonings will make psychological assessments that much easier.

      Reply

    • Kristin Blair
      Sep 17, 2021 @ 11:36:50

      Hi Bekah,

      You managed to make parts of your post relatable which really helps an audience get a more “real world” understanding of the topics if they can resonate with your examples. Hence, me enjoying your comparison with the rise in book sales at AU and the enrollment of female students.
      I also really loved how straightforward you presented your information. It made it easy to understand and easy to follow. Great post!

      -Kristin

      Reply

  7. Will Roche
    Sep 15, 2021 @ 11:32:13

    Throughout my undergraduate experiences, I have come to learn that correlation does not equal causation. First and foremost, I think that this is a common mistake in society because people tend to ignore the numerous confounds in many situations where they find a correlation. An example of this would be that seasonally, when the weather gets cold, people tend to spend more money at retail stores. Some people may automatically assume that the cold weather causes people to want to spend their money. However, people may be quick to forget that spending money on winter clothes may simply be more expensive than summer clothing, and that this uptick in spending may also be due to the holiday season.

    In terms of the mental health field, I think a primary issue with discerning correlation and causation may have to do with the current model of diagnosing people. The medical model seems to strive for causation, (certain symptoms mean you have this, if you have x it’s because x happened to you when you were a child). However, I think it’s important to note correlation over causation because of the increasingly strong movement of spectrum’s of diagnosis (just because x happened to you when you were younger, does not mean it caused you to have x diagnosis).

    I think there are many reasons why reliability is so important for psychological assessments. There are many implications for what a reliable assessment can mean for an individual. First, it’s crucial that the assessment is properly and repeatedly assessing what the therapist needs to determine for each client. If the therapist is administering an assessment for clients with anxiety, it is of utmost importance that the scale is reliable so that the therapist can make an assumption of severity for each client based on their answers/scores. If this test is unreliable, there’s potentially no rhyme or reason for why anyone received the scores that they did, and therefore could lead to multiple issues, such as wrongful diagnosis. Without a baseline reliability for a measuring instrument, therapists would not be able to find answers, and then therefore would not be able to find solutions for the clients, which is what makes reliability in measurement so important.

    Reply

    • Lauren Pereira
      Sep 15, 2021 @ 14:30:03

      Will,
      I also learned a lot about this in my undergraduate years and I find it significant to be brought up again in our graduate program. It puts emphasis on the true importance within correlation and causation and how they do not depend on one another. You bring up several good points, especially the fact that it can be easily mistaken in our society because of people ignoring these certain situations. You used a nice example to explain this.

      You make several good points when describing reliability. I agree that there would not be much reasoning without reliability because it is what we depend on when trying to make good decisions for our clients. Therapists would struggle with helping their clients find better outcomes for their situations. This has not only covered why reliability, alone, is so important, but it has also covered how important it is within psychological assessment.

      Reply

    • Madelyn Haas
      Sep 15, 2021 @ 19:22:38

      Hi Will,
      I enjoyed reading your response to the prompt. I agree that people assume correlation equals causation because people simply do not consider outside variables often. It is easier to just assume that A causes B instead of considering there maybe factor C that influences both. Also, I had never heard the example of winter and spending before, but I think it is a great way to illustrate your point. I also used a seasonal example (shark attacks and ice cream both increase during the summer) for my response.

      As for your response on reliability for psychological assessments, I appreciate how you went into detail with your example on anxiety. I made a similar comparison with depression and assessments. There is no reason to be giving assessments if they are not reliable. Clients should be our top priority as future counselors, so we should always check the reliability of assessments to make the most informed decisions for our clients.

      Reply

    • Tom Mandozzi
      Sep 15, 2021 @ 20:47:53

      Hi Will,

      I really appreciate the example you gave about spending money on retail in the w-Winter versus the Summer. I had not thought of the concept of correlation vs. causation in this way and I think its a great application of this concept and in showing that correlation does not mean causation. You identified two potential “variable Z’s”, that could impact this relationship: the cost of clothing might be higher for Winter clothing than Summer clothing, and people might spend more money in the Winter if they shop for the holidays. These were really helpful examples!

      I also completely agree with your points about the importance of reliability and how it informs treatment in the field of counseling. A therapist must trust the reliability of an assessment they are providing to their client in order to use the results as a tool for developing treatment planning moving forward. Great points here!

      Reply

  8. Lauren Pereira
    Sep 15, 2021 @ 14:08:48

    In psychology, it is common to associate correlation and causation but it is important to remember that one may not be the cause for the other. Causation goes deeper than correlation does. Correlation can be caused by a third party, not based completely on causation. The third party may be the reason for affecting both of these descriptions but in different ways. Individuals might make this simple mistake because of noticing that these two variables can be positively correlated, but this does not mean it is relatively associated.

    For example, the more frequently you eat ice cream does not associate with the more likely you are to get a sunburn even though they can be positively correlated. They may depend on the factor that it is summer and the weather is warmer. In this case, they do not depend on one another. It is important to realize that these two forms of reliability act independently which shows that they do not have to relate to one another and they do not cause another to become positive or negative.

    There is importance behind the term reliability, especially in psychology. This measure comes up a lot within assessments. It can be used to determine consistency which is significant within psychological assessments. This can help to clarify results within studies as well as fulfilling our clients needs when properly assessing them. These assessments need to be as accurate as possible and reliability is a much more dependable variable than anything else. If assessments can not be completely reliable, you may not be seeing any progress or changes within studies or assessments.

    Reply

    • Madelyn Haas
      Sep 15, 2021 @ 18:41:16

      Hi Lauren,
      I appreciate how in your response to the correlation/causation prompt you mentioned a third party that could potentially affect the preexisting variables. It’s always important to look out for outside variables that can influence others, especially in experimental settings. Also, coincidentally we used similar examples. You used ice cream and sunburns as your example for correlation not causation, and I used ice cream and shark attacks, haha.

      As for reliability, I agree with what you said completely. I believe that without reliability assessments serve no purpose in either experiments or in the treatment plan of clients. It’s important to consider that without reliability no measure would be valid.

      Reply

    • Tressa Novack
      Sep 16, 2021 @ 15:14:22

      Hi Lauren,
      Great post! You give a really great example to help explain correlation. We may eat more ice cream in the summer, and our time spent outside in the warmer months may increase our likelihood of sunburn, but one does not cause the other. In terms of reliability, I also pointed out how important it is in research, specifically when we need to draw conclusions from results. It is imperative that whatever instruments we use in research are reliable so that we know we can draw conclusions from our results. I also talked about how reliability is important when assessing out clients’ needs. The assessments we use in therapy must be reliable so that we can provide our clients with effective treatment.
      Tressa

      Reply

  9. Olgena
    Sep 15, 2021 @ 14:33:41

    The first time I heard about correlation concept was in my undergraduate stats class, and I could say that was one of my favorites. Even when I saw this week discussion prompt the immediate thought that came to my mind was this scheme: A is Not related to B! A is related to C! B is related to C which means that even though A is not related to B there could be a third factor (C) which might or not link them together. I think that the misconception that correlation equal causation happens not just in the mental health field, but is randomly used in most people daily conversations where without solid evidence we point out reasons as causes of a situation or an event happening. Especially in the mental field correlation is mistaken as causation because there could be variables that overlap with each other in different cases, but that does not necessarily mean that they are the cause of a specific “problem”. For example, taking one of the psychological myths which says that infants that have been exposed to classical music before they are born tend to be more intelligent. This fact does not mean that listening classical music causes increment of one’s intellectual performance. There could be a variety of other reasons related to a person performance on a specific task. During my undergraduate years, I also have read other studies describing how students who listened to classical music before a test performed better than those who listened heavy metal. Again, we could not say that the music type was the causation of their performance without knowing more about their academic background, social, psychological state etc. which could significantly affect their ability to perform in a test.

    As for our second question YES, I agree reliability is wicked because I think is so closely related to coherence which is a very important element when it comes to evaluate clients. When I started as a behavioral therapist, clinicians did not meet with the clients as much as they do now because of the pandemic. What I noticed during that time is that the treatment plans for our clients were not as successful and effective as they are now because I believe they were based in insufficient “information and observation time” to best evaluate someone’s behavior. There would be cases where the client maladaptive behavior would increase, because of a certain activity or method used in treatment plan instead of decreasing that behavior. This fact made me realize that treatment plans are not always fully accurate, and errors may occur. It is important, especially for those who work in mental health field to be aware that there are errors happening when it comes to measurement instruments and make sure to find the best possible way to find effective ways to decrease them. I think is very important to know what, who/which and how we are measuring and evaluating different variables, and pay close attention to ” error” as well. It could be is a significant factor that shows how effective and accurate is the instrument used on a specific case.

    Reply

    • Moises Chauca
      Sep 17, 2021 @ 00:25:59

      Hello Olgena,

      I really liked your example about correlation and causation. I also remember about A not related to B but B related to C from my psych stats class. Your example also reminded that correlation are not directional and that causation is. I also totally agree about your explanation on misconception of causation. One fun fact about me is that I did a research study on the effects of classical music and rock music on reading comprehension. I had to point out that there was a relationship between the variables and the other variables were in play like the examples you gave in your post. Lastly, I have to say wow! Good job explaining the importance of reliability. Your example was so good that it even help me understand the concept better. I really like how you incorporated your experience on the example.

      Reply

    • Moises Chauca
      Sep 17, 2021 @ 00:27:40

      Hello Olgena,
      I really liked your example about correlation and causation. I also remember about A not related to B but B related to C from my psych stats class. Your example also reminded that correlation are not directional and the causation is. I also totally agree about your explanation on misconception of causation. One fun fact about me is that I did a research study on the effects of classical music and rock music on reading comprehension. I had to point out that there was a relationship between the variables and the other variables were in play like the examples you gave in your post. I have to say wow! Good job explaining the importance of reliability. Your example was so good that it even help me understand the concept better. I really like how you incorporated your experience on the post.

      Reply

  10. Madelyn Haas
    Sep 15, 2021 @ 15:45:51

    While correlation does not equal causation, many people erroneously assume that it does. I believe that people make this assumption because they want to find a cause and effect for everything. People learn growing up that their actions cause things to happen. If that logic applies to a lot of things, then why wouldn’t it apply to everything? It, of course, does not apply to everything. Many things have correlational relationships without either causing the other. A common example of correlation but not causation is ice cream sales and shark attacks. As ice cream sales go up, so do shark attacks. That isn’t because one causes the other but because more people eat cold ice cream and go to the beach in the summer.

    People jump to correlation equals causation a lot in the field of mental health: A field that is still relatively new and full of nuances. Not only that, but the field of mental health is different from a lot of other scientific fields. With a field like medicine, there is often a straightforward cause and effect relationship (e.g., a certain type of bacteria causes a patient to experience vomiting and diarrhea). One theory to explain psychopathology is the medical model which takes a lot, as the name applies, from the field of medicine. While the model is applicable in a lot of situations, it does lend itself to people applying causes to correlational relationships. If substance abuse and depression are correlated, then does substance abuse cause depression? Does depression cause substance abuse? Is there a mysterious third factor that causes both? Realistically, people make jumps from correlation to causation because they want answers. If we found out that there was a third factor that causes both depression and substance abuse, for example, we could help a lot of people in treatment. Even though that sounds wonderful, it is best to be cautious and not jump to conclusions on causation.

    Reliability is extremely important for psychological assessments. If an assessment was not reliable, there would be no purpose for it. Imagine: Each time you take a personality test you get a different score/answer. What would that test be telling you? Nothing at all. Without reliability, psychological assessments would be pointless and would serve no purpose for your clients. Not only that, but unreliable assessments could actually cause harm. If a depression inventory wasn’t reliable, you could wrongly assume that a patient who is very depressed is getting better when in fact they are not. That could be dangerous for the patient and negatively affect your therapeutic relationship with them. Without reliability, assessments can never be valid and never measure what they are intended to measure.

    Reply

  11. Kelsey McGinness
    Sep 15, 2021 @ 19:39:24

    The assumption that correlation does not equal causation is a common mistake in the field of mental health for a variety of reasons. One of which being that it is often wrong to assume that even though two things reliably demonstrate a correlation with in another, it does not mean that one in turn causes the other to occur. For example, if a news report was to say that a decrease in sunglass sales is correlated with an increase rate of car accidents…it does not in fact mean that the lack of sunglasses bought by consumers influenced the increase in number of car accidents. These two things may have occurred around the same time but they are not related. The same thing goes for mental health disorders and environmental, social, or other variables which may occur simultaneously but may not deem to be the cause.

    Reliability is considered to be a critical indicator of consistency across assessments and data collection in the field of psychology. It is important for psychological assessments to have reliability so that assessors and psychologist can pick a fair and reliable assessments that have valid and repeated criterion the measured the desired behavior across a variety of people.

    Reply

  12. Tom M
    Sep 15, 2021 @ 20:19:48

    I believe that correlation is often strongly used to justify causation because, as the textbook explains, correlation is used to examine consistency. We as humans look for consistency and patterns and like to make sense of things using a “cause –> effect” lens. We want to believe that variable X effects variable Y because it is easier to make sense of the relationship between two variables than trying to make sense of additional factors. The lecture explained this point well; there is always another variable involved that is not known or measured (variable “Z”). Especially in the mental health field, there are often many factors that contribute to a specific behavior or set of behaviors. To build off the example in the textbook, we might expect a negative correlation between the amount of counseling sessions a client attends (variable X), and their level of depression (variable Y). Just because we observe a negative correlation between these two variables, does not mean there is not an additional variable (variable “Z”) involved in this relationship. Variable “Z” could be a change to a more effective medication to treat depressive symptoms, a change in living environment, etc. The amount of counseling sessions might correlate with a decrease in depression levels, but it does not indicate causation due to the potential for multiple additional variables. Therefore, I think the field of psychology, and more specifically counseling, is both parts intriguing and extremely difficult.

    In my work as a Therapeutic Youth Support through the Children’s Behavioral Initiative I commonly see children and their parents hoping for a “quick fix” to a perceived problem. They are often looking for the “answer” to changing or eliminating maladaptive behaviors. They often expect the clinician and I to give them these answers so that the problem can be “fixed”. We often must explain that there is not single “cause” for the desired “effect” but that there are many factors at play when it comes to looking at the broader picture of the behaviors. We know that when our cars run out of gas, we must refuel our gas tanks to solve this problem. There is a clear “answer” to fix the problem here. However, in counseling and psychology there is no clear linear relationship to getting from point A to point B. We can implement different interventions and revise our individualized action plans as we go, but we must consider multiple variables for behavioral presentation as there is always another variable that is not known or measurable.

    I think reliability is critically important for psychological assessments because we want to make sure we can trust the information an assessment is telling us. More importantly, we want the client to be able to trust the results we are implementing the assessment to determine. Assessment is important because it informs what interventions will be implemented and the treatment planning moving forward. If the assessment and gathering of information aspect of treatment is not reliable, this could hinder the implementation of appropriate treatment for mental health disorders. If interventions are implemented that are not appropriate in counseling, there could be more harm than good as a result for the client.

    Reply

  13. Teresia
    Sep 16, 2021 @ 12:17:59

    As a psychology major in undergrad one thing that was always stressed was correlation does not equal causation. In simpler terms this means that because two things correlate it does not mean that one caused the other. The correlation could be caused by a third factor that affects both of them. I think we as humans like to find an easy way to explain things, something happens (cause) that then makes something else happen (effect). One of my favorite examples that helped me understand this concept better is. Should ice crean be blamed for murder? There has been a correlation between the increase of ice cream sales and the increase in homicide. This example shows that correlation does not equal causation because there must be a third party factor like the fact that ice cream sales rise during the summer and that there are more people out gathering around the streets making finding a target easier.

    Reliability is important because it refers to the consistency of a measure. In psychological assessments reliability is wickedly important because clinicians use tests and they trust the test to be reliable and that they can get the same results. For example if a clinician is giving an assessment on neuroticism, each time it is taken the results should be the same. If the result drastically changes each time we learn nothing from it. Overall the reliability of an assessment is important because it helps clinicians choose an intervention method that is fitting based on the results.

    Reply

    • Kelsey McGinness
      Sep 17, 2021 @ 20:57:59

      Teresia, I like the example you provided for correlation does not equal causation. I agree as humans we look to find the simplest explanation as to why things occur despite it not always being the case, and as you mentioned there is often a third cause. I also like the point you make on reliability when you mentioned that if the results change from test to test it displays unreliable data and therefore can not be used for assessment.

      Reply

  14. Tressa Novack
    Sep 16, 2021 @ 14:18:13

    Correlation represents a relationship between two variables. In order for the variables to have a relationship, they must have some type of influence over one another. I think that is why so many people get correlation and causation confused, especially if the correlation is strong. I will make up an example. The correlation between not wearing sunscreen and then getting sunburn is positive 0.8 (according to me). This would indicate a really strong relationship between the two variables. I’m sure that most of us have gotten sunburn before because we did not put on sunscreen. Based on this strong relationship, a person might think that not wearing sunscreen is the cause of sunburn. However, I have been out many times in the sun without sunscreen and have not gotten burned. This could be because of the UV index, cloud cover, if I stayed in the shade, how susceptible my skin is to burning, and what I was wearing. With all these factors we can see that lack of sunscreen does not cause sunburn, but the two are certainly strongly related. I think that in the mental health field people may forget to consider all the factors that can influence the relationship between two variables. If we take a strong correlation at face value, it can be easy to say that one causes the other. However, we have to remember that there are other factors affecting the relationship between the variables.

    Reliability is so important in psychological testing, because it is needed before we can have validity, but also because we need consistency in testing. If we cannot rely on an instrument to test something consistently, then how can we accurately test for anything at all? If an instrument that is meant to measure depression gave a different answer every time the person took the instrument, then the counselor would not have an accurate idea of how depressed the person is. This would be detrimental to the client’s treatment, because a counselor would not be able to provide a treatment plan that would fit the client. Without reliability mental health professionals would struggle to come up with effective treatment plans. A lack of reliability would also negatively impact research, because researchers could not be sure that their instrument is accurate. This would affect results and the conclusions researchers draw from results, which impacts theory and practice in the mental health field.

    Reply

    • Monika Dhamale
      Sep 17, 2021 @ 18:07:05

      Hi Tressa,
      I really like how nicely you explained that correlation represents a relationship and how variables might influence one another without one variable causing the other one. The examples you provided makes it super easy to clear the confusion between correlation and causation. Also the point you mentioned about clinicians choosing intervention method based on test results helps understand how crucial reliability is especially in mental health assessment.

      Reply

  15. jeremy
    Sep 16, 2021 @ 15:20:25

    This mistake is still so especially common as we naturally try to tell a story, Casual knowledge of statistics would make this distinction particularly weak leading to confusion. The common mistake when looking at correlation is people assume all confounding variables are controlled for, there may be several reasons why time spent online and antisocial behaviors are both on the rise. Human behavior is messy, and it is hard to make a test that can prove causation and still be broadly applicable. Thus we largely use correlational research in the mental health field.
    Reliability is important for psychological assessment because these tests are at the cornerstone of therapy clinical aspect, for Clients with insurance, it is required to have evidence of a disorder in order to continue treatment. Therefore a consistent test is nessacary to ensure that there is a set instrument, not only to diagnose clients but then to track progress over their time in therapy. Tests with stronger reliability are better at providing a measure closer to a client’s true score, meaning that clients are denied necessary coverage less, and their progress can be tracked over time.

    Reply

  16. Alexis Grey
    Sep 16, 2021 @ 16:24:57

    Correlation does not equal causation – and in fact what may appear to be a causal relationship based on the relationship between variables alone may not be at all. Correlations simply represent the relationships two variables have but it does not explain why that relationship exists.

    An example from my past studies that always stuck with me regarding this concept is that there is a positive correlation between ice cream sales and drownings each year. Does ice cream cause drownings? No of course not – both instances of drowning and ice cream purchases go up in a linear type way due to the summer season. Why is the assumption that correlation equals causation still a belief? If I was to guess I would say that there is just a lack of literacy in the area of statistics. At least in my experience – I was not introduced to the concept until very late in undergraduate, and even then, it was not something that was drilled into my understanding until I took research methods at the graduate level.

    I also think when you are looking at the nice clean visual created by a strong positive or negative correlation – it just FEELS like a causal relationship. I don’t know how to explain that better but you see a graph like that and in your mind, you tend to say “Ah yes – those two things have a relationship, with one comes the other – one must cause the other”. Maybe this just relates to the human tendency to see patterns in relationships and assume cause – not sure. Someone help me articulate.

    Reliability is important in psychological testing and testing in general because it means we wouldn’t get – or should not expect to get radically different results if we give the same assessment to the same person. So, you should not be getting an IQ of 89 one week and 135 the next. A test that unreliable doesn’t mean anything. Results from psychological tests and measurements are used and considered when making big decisions regarding that person’s treatment or mental health. So of course, it is EXTREMELY important that the tools we are using to reach conclusions about care are reliable. If a child in a school system is held back or removed from regular classes due to a low IQ score on a measure that is not reliable, then that child is going to miss out on a normal and appropriate school experience due to a measure of intelligence that may OR may not be accurate to their intellect and academic potential. Whenever client or patient care is involved reliability of the tools used in those decisions is so WICKED important.

    Reply

    • Moises Chauca
      Sep 16, 2021 @ 22:36:49

      Hello Alexis,
      I enjoyed your explanation about correlation and causation. I totally agree with you on visual of strong correlated relationship can be seen as causal. I can add that it seems like that, especially if it makes sense. Like the example I used about aggression and violence on TV, if you look at it from an outside perspective, it would make total sense if they cause each other. I also agree with your examples of reliability because an assessment that is giving you different scores on the same test is not good. Reliability takes such an important role in assessments that if there is more error than consistency it can lead to many problems, just like you point out with your example about a child in the school system. Lastly, I like how you used uppercase words to make your point across.

      Reply

  17. Sandra Karic
    Sep 16, 2021 @ 17:37:27

    I feel like the tendency to confuse correlation for causation might be related to human nature. I was reminded of our discussion in abnormal psych about classical conditioning, operant conditioning, observational learning, and how humans are more likely to imitate a behavior if it seems to produce favorable consequences. I think a lot of the times when we confuse correlation with causation we have limited information and are making assumptions based on what we can observe with the goal of gaining more information and potentially avoiding a negative consequence or increasing our likelihood of a positive consequence. I think to a degree we might be wired to make assumptions based on observations which relates to how likely people are to believe anecdotal evidence. If you see A happen right before B most of the time, it’s a lot easier to blame A for B instead of trying to find a third variable. I think the conflating of correlation with causation happens frequently in the field of mental health because there are almost never any clear cut answers and humans really don’t like that. Additionally, there’s a lot of pressure to find clear cut answers, partly due to the medical model’s assumptions on organic causes, and partly because of financial incentives. A journalist who describes a scientific study by saying Violent Video Games Cause Violence is probably going to get more exposure to the public than a journalist who writes the headline Violent Video Games Related In Unknown Way to Violence. I also thought of the concept of neurorealism, where people are particularly prone to confusing correlation and causation when it comes to brain imaging studies. One study found a correlation between teens playing violent video games and decreased brain activity when shown violent pictures, compared to teens who did not play violent video games. A lot of people interpreted the study as saying that video games were desensitizing teenagers to violence. However, when the study was repeated using colored violent photographs, instead of black and white violent photographs, the difference in brain activity between the video game and non-video game groups was no longer observed.

    I think reliability is wicked important in determining the efficacy of an instrument. If a test on introversion produced different results in the same person each time, it would be pretty difficult to make a meaningful interpretation of the results. Impossible really. The same could be said about an unreliable intelligence instrument, except while an unreliable measure of introversion may not cause direct harm to the subject, an unreliable instrument that is supposed to measure intelligence could prevent a child from receiving vital support and accommodations. If an instrument lacks reliability I don’t know how you would be able to interpret the results, and if you can’t interpret the results, what’s the point of using the instrument?

    Reply

  18. Pilar Betts
    Sep 16, 2021 @ 18:27:46

    1. The statement correlation does not equal causation is important especially in the health fields. Data can be misleading so it is important to look at the context closely. I wrote a paper once about a misleading statistic I found on social media, the graph depicted the amount of people in prison going up and the amount of people in the mental hospitals going down. Looking at the graph one would think that people with mental illnesses are being placed in prison rather than in institutions so that they can get the proper help they need. While this is a very real issue that does happen, this particular graph was taken from a study and the actual study was showing the relationship between people in prison and mental hospitals and homicide rates. The study wanted to express that people with mental illnesses are more likely to be victims of violent crimes. Looking at the graph without the additional context the correlation between the imprisonment rates and people institutionalized would be that when the amount of people in prisons increases the amount of people in mental institutions decreases. However, this is not the cause of the increase in prison population. Correlation can be used with anything, it’s always about proving a point. Graphs like the one I discussed are used all the time in media and politics. To use a simpler example of why correlation doesn’t equal causation, just because the number of yellow popsicles at the store decreases on wednesdays and the number of seniors in the store increases doesn’t mean the cause of the yellow popsicle shortage is the seniors buying them, maybe there is an additional factor like on wednesdays there is a sale on yellow popsicles. Anyone can make a graph with two variables showing either negative or positive correlations but that does not mean that variable A causes a change in B.

    This is important in the mental health field because as one symptom increases, for example increased anxiety and the person’s diet also has changed, that doesn’t mean the change in diet is increasing this person’s anxiety. A week later the client reveals they joined a new club at school, their anxiety increased because of their social anxieties about joining the new club. It is important that we don’t just run with a possible correlation in this field because if we do then we may provide treatment for something that is not the actual issue. Additionally in mental health research it is important that we factor in error because mistakes are made, that is why multiple trials are necessary and consideration of confounding variables. In Abnormal we learned about General Paresis and people believed it was caused by the jerk of train cars because it was travelling salesmen who most often had it., This is the perfect example of correlation not equalling causation. General Paresis was actually caused by untreated Syphilis and therefore it makes more sense why travelling salesmen got it so often they probably were having sex in all the different places they travelled and were more likely to be infected because of that.

    2. Reliability is important for psychological assessments because it helps to ensure that data is as consistent as possible, consistent data means that whatever instrument is being utilized is working properly and efficiently. Also data that is consistent is helpful when you are trying to prove a hypothesis if for example the study is on the effect of caffeine and improper diet on depression and data shows that participants with a poor diet who drink a lot of caffeine reported feeling depressed more often than people with a healthy diet who avoided caffeine your data shows a pattern and therefore is consistent and your hypothesis now can be proven.
    You can test the consistency of data by calculating the correlation coefficient, the closer the coefficient to negative one or positive one the stronger your correlation and the closer to zero the weaker your correlation is. We want a strong correlation of data to demonstrate that two or more variables are related in some way. If the correlation is weak or has no correlation then the two variables don’t affect one another. Reliable data is super helpful because it can lead to discoveries of new treatments and techniques. It allows for progress in research about a variety of different issues and areas. It is also important that data is reliable in case an experiment is repeated, similar results should appear and then we can make new hypotheses about similar data. For example the data collected in many studies throughout the history of psychology led to further discoveries and allowed for the experiments to be replicated and adapted for different populations.

    Reply

    • Kelsey McGinness
      Sep 17, 2021 @ 21:01:49

      Pilar I like your explanation of reliability. When you bring up the point of calculating the correlation coefficient to determine the reliability across assessments from test to test. This is a point I did not think about when reviewing reliability, and your point helped me to better understand how to numerically determine reliability. Great points!

      Reply

  19. Moises Chauca
    Sep 16, 2021 @ 22:08:11

    Correlation does not mean causation because there are more variables that are in play when two data sets are related to each other. In the mental health field, people mistake correlation for causation because people see the relationship between the two data sets as an explanation rather than a relation. In some circumstances, it is easy to make the assumption about causation specifically if the data sets have a strong relationship, but it is important to differentiate causation and correlation because there are many outside variables that affect the relationship. An example of a correlation that I hear people think is causation is aggressive behavior increases with violence on television. From an outside perspective, it makes sense that aggression increase when you see more violence. However, aggression can be influenced by a person’s environment and/or socioeconomic status.

    Reliability is the consistency of measurements in a psychological assessment that is repeated on a group sample. Reliability is really important and needed because it shows which measurements of the assessment are consistent and which could be influenced by error. We need psychological assessment to be reliable and have more consistency than error, so the individual is assessed properly. Lastly, reliability helps clinical choose the best of the best psychological assessment for their patients because psychological assessment with low reliability causes many problems like wrong diagnoses or misinterpretations of scores.

    Reply

  20. Monika Dhamale
    Sep 17, 2021 @ 17:55:34

    Correlation does not equal causation!
    From my understanding this means just because two things are correlated does not necessarily mean one causes the other. Like from the example given in class by Dr. V, exposure to TV violence might lead to aggression or having aggressive tendencies is the reason why one watches violence on TV. While this might be the case there might be other factors affecting both of these variables. Another reason why correlation does not imply causation is that the sample we’re looking at might not be representative of the population of interest.

    One of the reasons why this is still mistaken in mental health field is, in correlational research, we identify patterns of relationships, but we usually cannot infer what causes what. Like from the example given in class, it cannot be taken to indicate that viewing violent television causes aggressive behavior. Although it is tempting to assume that viewing violent television causes aggressive play, there are other possibilities.

    When it comes to research especially in psychology where we have to measure unobservable psychological constructs like intelligence, it’s important that the experiments are credible. Reliability is important because we need to be sure that the findings will be the same if the experiment was done again.
    If the test can’t produce consistent results, then it’s questionable if the test is accurately measuring what it is supposed to be. Tests like GRE are used to make college decisions which have an important effect on people’s lives. Here, the test’s reliability has important consequence for the quality of decision made on basis on individual’s test scores. It’s difficult to trust that the data produced by the measure is an accurate picture of the participant’s performance if the measure’s reliability is poor.

    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 72 other followers

%d bloggers like this: