Psychology week 7 assignment

end of the course. I have decided, or I'm not planning to have a live session next week to kind of give you guys a night off. There's only a discussion next week. So I guess, you know, we're getting nearing the end. So I can't quite say congratulations yet. But next week, there is It's just a discussion, and so, you know, if anybody has anything they want to meet about or anybody has any issues with not meeting next week, let me know, but there is just this one discussion next week, no assignment, so that's, you know, that's all we're going to be doing next week. Does anybody have any questions about that or have a dying desire to meet again next week? Okay, I can't see any posts in the discussion. So, all right, so let's just jump in and talk about what we're going to do tonight because we do have quite a bit we're doing tonight and this week. um this week your discussion is focused on factor analysis and um that's one way that we can demonstrate construct validity of an instrument so we've been talking about a lot of different types of validity and if an instrument has some scales then we should be able to back it up with statistics and factor analysis, the way that we can do that. So reliability, we think about as consistency and validity, we think about as, you know, measuring what we intended to measure. But when we have something like a self-esteem measure, like we've been dealing with, we might have a measure that just assesses self-esteem as a whole or there might be subscales like different parts um you know in that same measurement tool so i'm sure you've probably seen in your reading some different instruments where they do have more than one factor. So that's what we're going to be focusing on this week. So I'm going to pull up a discussion so we can talk about that first. Okay. So for your discussion this week, your challenge, I saved a little spot here in case I wanted to add some kind of note, but your assignment is to give a brief discussion definition of factor analysis. So two sentences at the most. So I want you to be brief, but concise and clear. So with only two sentences, that means you've got to both cover, you know, what it is and just be, you know, pretty clear about it because they're just two sentences. For part two, I want you to find an example of an assessment instrument that has clearly delineated factors. And these are often referred to as subscales in the literature. However, if you find an instrument that has subscales, it should be supported statistically by a factor analysis. So some of you may even have instruments that you're considering that you talked about in week five or week six that have a factor analysis. But if not, I'd like you to share some other instrument that has um clearly delineated factors that's backed up by a factor analysis um and then um i would like you to um talk about for part three of this this discussion what the difference is between exploratory and confirmatory factor analysis so to do that um you can refer to your readings. Your readings for this week include chapter 10, as well as this article by Bryant and Yarnold that goes over principal components analysis and exploratory and confirmatory factor analysis. So this one definitely helps you with answering question three. um but you know your your book is very useful as well so um you know i want you to take your readings from this week and summarize the difference between exploratory and confirmatory factor analysis so it's part looking at your readings and partly looking at an original research study that has you know an instrument with clearly delineated factors that are backed up by factor analysis. So what is factor analysis? So factor analysis is taking a whole bunch of data and reducing it to smaller or shorter elements. I'm going to stop sharing for a minute, but that's what we're going to work on tonight. And we're also going to, you know, try to understand what some of the output that we have from factor analysis, what that's dealing with. So for example, you know, there are different factors. We're also going to try to interpret the rotation. And, you know, basically we're going to be, you take a large set of variables in a factor analysis and you reduce it into fewer summary parts. So with the principal component factor analysis, that's looking for patterns in items and seeing if they hang together. And then you can group into components and summarize that information. So one of the things we'll see when we looked at the output for a factor analysis is something called eigenvector, and that's like a mathematical term. But, you know, like a vector is a mathematical term. But this is basically where we're taking the broad data and we're condensing it and we're finding different underlying components. And each eigenvector tells us how much each item contributes. And so we look for eigenvectors greater than one, because if they're less than one, that means it really only is to taking into account one item. And if we only have one item, we're not going to have a factor or a subscale that's only one item. So the eigenvector is telling us how much variance is contributed. So we look for eigenvectors greater than one or, you know, you know, even better if it's greater than two, then it's probably accounting for even more, you know, of that data. Okay, so let's take a look. I'm going to share my screen again, and let's look at the assignment, and then I will go through an example of how to do the factor analysis. So for your factor analysis assignment, you're going to use the same data set that you had in week four, the Sympathetic Magic Scale, and conduct an exploratory factor analysis. Now, if you're using SPSS, the directions are analyze dimension reduction factor, and you will choose these different parts, which I will demonstrate. However, I want to point out, because some people were confused about this, who use Jamovi. Whether you're using Jamovi or whether you're using SPSS, use the same data file. So I think some people thought that there was like a Jamovi data file and that's, there's not really any such thing. Like with Jamovi, you can open like an Excel file or you open an SPSS file or something like that. But it doesn't matter whether you're using SPSS or Jamovi, you just, you open up the SPSS data file. So you'd go back to your data set. um i think somebody's speaking i'm gonna try to like mute if anybody's i don't know everybody please mute uh unless you're trying to talk of course okay so um again just make sure you're opening up the svss data file and then you're going to go through these different steps and this is what we're going to try to do tonight first first we're going to answer how many factors were extracted? Do we have just one like single self-esteem measure or do we have multiple components or multiple subscales in our self-esteem measure? And you'll do the same thing with your sympathetic magic scale data set. Number two, list the factors in a table. So if you have whether you have one factor or two factors or three factors or four factors or five factors, list what those are out and put them in a table. Part three, can you subjectively make sense of how the items load on each factor? In other words, what does each factor mean? um so I think I've mentioned this before when I worked at um well I may not have mentioned it but when I worked at Scholastic we did a lot of research with teachers and with parents and um I remember that one time we we subcontracted with a company who did research with for us with parents and they ended up, you know, having parents ask, answer all these different questions. And then they, they found parents to fall into certain buckets and they came up with names for them. Like what's kind of a name you might think of a, of a type of parenting style. Does anybody have, have a, have you heard of any terms or can you think of any terms for different types of parenting styles? Well, there's the unofficial helicopter parent. Right. So there might be a helicopter parent, might be one type, what might be another type. So maybe if some of your factors kind of fell really high in like, you know, need for control or supervision, I don't know, actually all the components that go into helicopter parenting, but really, you know, very involved that might be like oh maybe we would call them you know it's not going to give you a name you're gonna you would come up subjectively with a name for what that grouping is and maybe you'd say oh that's a helicopter parent okay so what what's something else katherine is that do you have your hands raised i think i have to take you off of mute yes authoritarian authoritative parent, permissive. So those are, yes, those are common parenting styles. So those, you know, I have a lot of students that have been doing research with that because that really follows like Ainsworth's attachment parenting style. I think what Melissa said was a little bit more of like parenting types. So you'd probably go with one or the other. I would go with like attachment like parenting styles then and that's a mixture of like warmth control and um things like that whereas some might be more to do with like at scholastic we are of course more concerned about education right so um like maybe the helicopter parent is like monitoring their kids gradebook all the time and keeping track of their schedule and helping them to figure out everything, even if they're in high school or college. You know, maybe there are some other types of parents. I remember, I don't remember the names of these, but I remember one was like the parents who, because of course, mostly we were dealing with parents of younger kids, like buy all wood products, like they wouldn't have like plastic toys. They would have be much more into like the natural, like maybe low media use, but maybe more like, you know, natural environments. I don't know what we would call those Johnny Appleseed parents, holistic. Yeah, maybe something like that. So you kind of get the idea that, you know, maybe if your items fall into like one of these buckets, you could kind of call them something like that. So it's somewhat subjective, but you can, you know, it's kind of just using your creativity. Yeah, BPA-free parents. Okay, so that's kind of fun. Part three is where you subjectively try to make sense of how the item's loaded on each factor, and you can come up with a fun name if you want, or it can be, you know, more academic. That's up to you. Question four, after examining the data, how many factors would you retain in your final solution? So you might see in your factor analysis what looks like, I don't know, you know, like five or six factors, or maybe even seven factors. But that doesn't mean that your final solution is going to be, I'm going to keep seven factors, you know, some of those factors might not be very strong, they might be very weak. and so you're better off just reducing it down to like three or four factors so explain your reasoning for including or excluding each factor and then part five try your best at writing up the result section so obviously this doesn't have to be perfect part five here just kind of try your best and you know try to try to go through the different steps so um so that's just basically an overview as i mentioned you're going to use the same database from week four so before i go through the example i'm going to demonstrate with our self-esteem data set do you guys have any questions before i before i do a demo Sorry, I'm still sharing my screen, but I was trying to see here. Catherine, I see you have your hand raised, but that might be from earlier. Any questions or comments before we move on? Okay. All right. Okay. So I'm going to share our self-esteem measure. so we we had the self-esteem measure which has id number age *** and then 15 items and if you'll recall we tested out 13 items that we then reduced um you know i i i think i was just checking it out before class and i think we were unsure whether we were going to keep 10 or 11 items but I just was trying it out and it seemed like if we took out these three items I am a good human I take constructive criticism as a helpful tool and that I feel proud of my achievements then we get all the way up to 0.892 So we still have 10 items left and we get pretty close to 0.9, which is excellent. So that seemed like a good place to be in terms of doing a factor analysis. I think we talked about last time that we could kind of keep going and maybe trim it even further. but since we want to see if there's more than one subscale, it can be useful to have a good amount of items. So in this case, we have 10. I think in your case, you guys have many, you know, many more than that. So I'm going to go ahead and show you how to do this. So we go to Analyze, dimension reduction factor and then we can go in and we choose those first 13 items now we didn't really keep 13 items though right we decided to remove three items so So we should be removing those that we determined to remove. So let me just double check which ones we said we would remove. I'm a good human. I take constructive criticism. Okay, so I'm going to take that out. So do the same thing. The ones that was determined that you should remove in week four should also be removed. So you should make sure I have those. I feel proud of my achievements. You should make sure that you're working with the reduced data set. So I think in week four, if anybody's unsure, if anybody didn't get it all right and is unsure, you can double check with me. but I think it was items 16, 17, 24, and 26. And you should be removing those items because they're no longer part of the scale because they didn't have good internal consistency. So you would put in those remaining items. so okay in this case we're going to be doing our principal factor principal access factor analysis and you should be following exactly as is shown here so for the extraction you should be clicking on the scree plot for your display and you unclick the unrotated um and then we're going to click okay for rotation we are going to click on very max um and this is because we wanted the rotated solution which is already checked here and so that's going to contents the individual factors into more definable factors so that'll make it easier for us to interpret. So, you know, you should just, you know, follow the instructions that are in the classroom, but this is the kind of, this is the kind of rotation that I'm able to do with the version of SPSS that I have. So, so then now we have our 10 items in here and I can click okay and we'll see what our output shows okay so I'm just just gonna double count check here we have five six seven eight we have ten items so we can see our ten items listed right here at the top and then we are going to try to figure out how many factors we have and we're only gonna to include or consider those that have an eigenvalue of one or greater. First, we can see from looking at this total variance explained, and we can look at the percent of variance and the cumulative percent of variance, and we can see that it says component 1, which would be our first factor, is accounting for 56.6% of the variance, which is a lot. That's a lot. And we can see that in the scree plot, right? So in the scree plot, our first factor has an eigenvalue between 5 and 6. So that's counting for a lot of the variance. So we definitely have at least one factor, right? That's very high up. Then if I look at the percent of variance that it says the second component has, it says 19.5. So the cumulative, that's three-fourths, right? 76%. So that also indicates a good amount of the variance. And then our third component, it says 15.6 or 90, cumulative 91.6. So I'm going to stop looking after factor three right here for now because it looks like we do have three, at least three factors. But let's look at the scree plot because that really helps us see what's going on. So remember I said we want an eigenvalue greater than one. So once we go to, you know, component or factor four, five, six, seven, eight, nine, those are all less than one for the eigenvalue here. So they really, you know, are not indicating that there's clear factors here. So it does show that we have three factors that are all with the eigenvalue above one, this line right here. Um, and if we, if we, what's really important to look at is our component matrix next. So, um, what, what we see here, um, well, I guess before we go, before we go through this, does anybody have questions about, I'm going to go through the rotated component matrix, which is going to tell us where each item falls. But before we go through to that, does anybody have questions about this SCREEP plot? The main takeaway here is that we're kind of looking at this cutoff of one and that we see our first three components, which account for over 90 percent of the variance in the data. these first three factors are really accounting for a lot of the variance and the rest of the components or factors are almost indistinguishable like very little variance they're less than five percent so those are really not meaningful and so so at this point we would just want to look at those first three factors or components and see which ones we want to retain and whether the variance that's being added is meaningful. Okay, so if we look at our first component, we would look at this first column, and we would see which ones have high values here. so um here i can see for example that i feel confident in my ability to handle challenges in my life that really falls into component or factor one so it has a very high value here of 0.985 um it's very minimal you know for point for factor two and factor three so we would we would consider this one to be part of factor one um whereas our next item i believe that i'm a person of worth at least as much as anyone else is is falling into factor two because it has a value you know we're looking at the close the closer it is to one so it's like we see that it has a value of 0.964. So that one falls into the second component. The third item, Expect Myself As I Am, falls into the third factor or component because it has a value of 0.894. um i feel that i have good qualities the highest value is with factor one at 0 .967 and that's really close to one um i believe that i'm a person of value um falls into factor or component three um i am deserving of respect is high for factor one at 0.834 but this one is interesting because it does have over 0.5 for factor three so it could potentially be part of factor three but we're going to take where it is the strongest and it's it's stronger or hanging together more with the other factor one items um likewise i can learn and grow from mistakes we see the highest value with um factor one for i can learn and grow from my mistakes um i believe my opinions and ideas are valuable again that's also factor one i believe i can accomplish anything i set my mind to. That's factor two. And then I trust myself to make important life decisions. That's factor one. So we can see that it looks like factor one has one, two, three, four, five, six items. Factor two, it looks like has two items, and factor three has two items. So it makes sense that factor one was taking up, you know, the most of the variance and is highest in eigenvalue because it has six items, right? So it was accounting for 56.6% of the variance, whereas component two and component three that each had two items are taking up about 19% and 16% of the variance. So that's kind of interesting. It's showing us that, you know, we can account for about 90% of the variance, but we have three, you know, three potential factors here that we're able to see in terms of our component analysis. So now that we know that, we can take a little bit more look at the items and try to make sense of it. So I will go through this a little bit more and then we can kind of I think we can we can also take a look at Jamovi as well. But before we before we do that, I want to just, you know, kind of finish this up in terms of trying to interpret it. But so I'm going to go through and like kind of where we can talk about, you know, about these three components. All right. So this is not a full write up, but I just pulled out the items just so that we could at least talk about what we might think those those items kind of represent or stand for. okay so what do you think our first component these are the items that fell together and so let's try to come up with a name or you know that we might want to call component one I feel confident in my ability to handle challenges in life. I feel that I have good qualities. I can learn and grow from my mistakes. I believe my opinions and ideas are valuable. I trust myself to make important life decisions, and I am deserving of self-respect. What do you think those six items might be representing? Does anybody have an idea of what we could call that? Self-confident. Yeah, I think that would be a good name. We call it self-confident. What else do anybody else have any ideas of what we might i mean all of them but the last one could be self -efficacy yeah i would agree with that too yeah so sometimes the name isn't going to be perfect but you might we might call it like self-confidence slash self-efficacy or something like that i think those those are good um great great um so component two i said i believe i'm a person of worth at least as much as anyone else i believe i can accomplish anything i set my mind to oh i wrote global self -worth so i should have taken that off but does anybody have any other name you might want to call it maybe self-concept good work um nice and then um component three maybe it's self-acceptance or maybe you guys have some other name for it like i accept myself as i am i believe that i'm a person of value It could be self-value. I mean, just using the same word as the item. Yeah, that's good. Self-acceptance works too. Yeah. Great. Okay. So that kind of just shows you an example of, you know, in this particular case, I think partly because we have a small number of items, you know, know, you really, it really is probably not too, you know, you, it's really not too surprising to only have two or three factors, but I've even done studies where we've only ended up with one factor. So this, this was at least a little interesting. It showed more than one factor and that there were some different subcomponents um so um let me see let me try to go into um jimobi and we can see yeah self-value or identity yeah that could be too so i'm sure we could we could probably spend some more time on that coming up with some fun names um Um, so, oh, great. I just lost my Jamovi data set. Well, that's, that's the thing I think I've, I've mentioned before, if you don't pay for it, um, you know, it does eventually go idle or sometimes it is busy. Um, and I don't have a paid subscription to Jamovi, so I have to bear with me. Hopefully it'll end up going through, but while we're waiting for it, how is it different? I know somebody mentioned in the discussion area that it was different. What did you notice is different about it? Yep, it was me, Dr. Kelly. So I was playing around with it, and the setup is different. It's different. Okay. Well, I think I've said this before, too, like when, you know, you guys are, I guess, you know, really learning with me because when we use Jamovi, because I obviously have a lot of statistics experience. I have a, you know, my minor concentration was in quantitative methods. So I've taken a lot of stats class. I've done lots of statistics in my research over the years. Um, but I've, and I've used a lot of different statistical programs, so I'm not afraid of any statistical programs, but that being said, I just, you know, we recently just started allowing people to use, um, Jamovi, so I'm just less experienced with it. Once you go into that factor analysis, the exploratory factor analysis, once you go in there and then you pull everything over except for well yeah whichever ones we want yeah we yeah let me see and the extraction under method after you get that done goes to the principal access okay so let's take out i'm going to take out two and three i don't know if that's right is that right oh no i i might be taking out more than i should okay i'm gonna have to go back and check which because now i have nine three four five six seven eight nine okay so one of these has to come back um okay yeah okay i'm just trying to see okay i think we need are there other people using jimobi oh yeah you're not the only one don't worry don't worry you're not the only one this is the first time i've used it so oh yeah as the ss ran out so i'm like well that's as good as reason as any um i actually think it's kind of cool some of the things that it does so um i'm definitely open to using it so i'll kind of share what i did and then maybe i mean i don't know if you don't know yeah um I mean I think here like here I just have nine items so right now it's only showing two factors but again I don't have I'm missing one of the items so I'm trying to the one thing I don't like about Jamovi is I can't see the item names as easily so it's a little bit harder for me to tell. Well, that's one of the reasons I said it looks different. Ah, yes. So it looks like I took out number nine, 10, 11, and 12, nine and 12. Okay. So I got to put 11 back in. Okay. okay so um let's just see hopefully it gives us three factors here too but oh yeah it does so so here i think that partly what it's doing differently is um if you're doing it But in SPSS, it was giving me the principal. Well, we're not done over here under method. Right. If you hit under that little tab under method and put in principal access. Yeah. And then do Baramax right there. yeah and so now i also hit um go down yeah the base on flat agent value i'm not sure if i'm saying that correctly did you do we need to put that under factors the number of factors over there on the left um so we want eigenvalue greater than one that's what i did and then i hit factor summary under additional output and then initial eigenvalues above screen plot i'm not sure if that's what i was supposed to hit but that's what i hit to see i think that's fine i mean you definitely need varimax and you definitely need scree plot um the issue is okay so it's interesting because it's showing here it's showing three factors but then once we go into trying to see yeah so the screen plot it's it's looking like I mean factor two looks like it's right at about an eigenvalue of one but it's it's different so i guess this is this is partly why it's good for you guys to see oh i have a couple more things clicked under assumption checks are we not supposed to i have the camera no you don't have to have those i have a question is it um it's supposed to be principle component analysis well that's well that's what i was just gonna say is that we're to get a different there's these different types of analyses here and so you're going to potentially get different results if you do different types of yeah analysis so um principal you that you have a very max rotation right here yeah so i thought we were supposed to do the exploratory correct no pca i think yeah so let's go back to the pca because that way principal component analysis because that way it is at least the same thing as what's being done in SPSS because otherwise you guys are going to end up with different results, um, depending on if you use SPSS or Jamovi, which we don't, we don't want to have happen. Um, I'm just trying to see here, which ones I took out. So I'm very glad you did this one then. So just make sure you're putting in, that's why I keep double checking myself, make sure you're putting in the right items. So what did I say? The analysis has been canceled. that's not good um so i don't know if this is because i'm using a free version let me try to refresh it um but yes we let's try let's all do the i mean it let's all do the principal component analysis because that's what at least the base or the, you know, version of SPSS that we have, at least that the university has, will do. So that way we're at least all doing the same type of factor analysis. So you can do the principal component analysis, but be sure to discuss in the discussion area what the difference is between these different types of factor analyses okay so i think i'm going to do is let's see i'm going to make sure you pick up the screen plot because we do want to see the screen plot um Um, and it's interesting because when I did it, my, um, factor analysis in SPSS with only nine items, it also had just two components. Um, but, um, it's interesting that with, with the same 10 items, it would, it would show something different, but let's look at the screen plot. So I think maybe I don't know why it was showing like a second line here. Okay, so this is what we want. We want to make sure we have this clicked based on eigenvalue, not rather than the not the parallel analysis. So you want to make sure that your rotation is varimax based on The eigenvalue is greater than one and you have scree plot checked off. And then then it looks like our we still have three factors here. Let me scroll up and see what it says. Looks like if I scroll up too far, it gets very fussy. OK, so. All right. It doesn't like me moving my screen too fast, but it looks, I can't really keep it. That's like, I don't know. It's like, it's hard to, I don't know why it's, some of it's not showing, But when I look at this, it looks like there are potentially three components, but it doesn't like item 11. So item 11, it's saying like, this is not good. It's all negative. Right. So that's saying, I don't like that. That shouldn't be part of my, my factor here. So let's try it. Let's try a principal component analysis without item 11 and see if that makes a clearer factor analysis, because it's saying there's something problematic about that because it's all negative. And so that doesn't make any sense. So let's see. okay so if I go down to that then I end up with nine items and it looks like they're okay although I don't know why the second one is negative but it does at least show again I don't know why it's sorry that my screen it just uh it doesn't like to stay in it, but then it shows me like two components, which if I took out, um, if I took out, um, number 11, okay, where is it here? So if I took out number 11 in my version of SPSS, I'm sure it would say this, I'm sure it would do the same thing. So sometimes you have, maybe you find something that's problematic. And so, you know, maybe it says like, well, we really don't want to have, you know, all of these, like, if I take out my number 11 which was I believe this one yeah let me see so I'm going to cross compare I feel confident in my ability let me see I'm a good human I believe I can accomplish things I set my mind to okay so we have taken out three nine and twelve and then it was saying i trust myself to make important life decisions okay so if i take that one out and i am left with these let's see what happens here so i think what this is saying here is very interesting. So if I just have those nine items and I do my reliability analysis on those, so I take out item 11, I take out item 12, I take out item three. And I know that I am a good human. It didn't like that one. I think this, let's see if this was, how many of these was, it still has 10 in here. I take constructive criticism. Oh yeah, that's the other one. I take constructive criticism. So if I take constructive criticism, then let's see how that ends up with our alpha. So, yeah, it does give us a stronger Chromebex alpha, 0.912. Again, I kind of was resisting taking out too many items because I wanted to check it out with our factor analysis. but we certainly could take out that and still have a strong, a strong set. And in fact, it looks like if we took out, I believe that I'm a person of worth, we could have an even stronger scale. So I don't want you to have to go back and do more things with your scale, because you guys already did that but i'm just showing you this just because i think sometimes when you when you go through um factor analysis you can find that there's even other items that that you could pull out and have an even stronger set this is actually kind of crazy because what this is really showing me is that with the items that we came up with for our self-esteem measure that we we could have a pretty short instrument like we don't need to have that long of an instrument because a lot of these items just kind of hang together and we don't need we don't need them all like we could really have like a five question survey. But at the same time, like I think with this, depending on how many items we end up having, we could have two or three factors. And, you know, I'm just going to, I'm going to check the sympathetic magic scale for both SVSS and Jamovi so that if there are different results depending on how you do it that you know you won't be penalized for that obviously we wanted to set it up so that your grade is not dependent on which software you use like that really doesn't seem necessary certainly to have to buy SVSS or rent SVSS just because, you know, because we only use it for two assignments in this class. So go ahead and use whichever one you're more comfortable with. The results should be pretty similar. I think in this case, it's just, it was a little fuzzy whether there's two or three factors because some of those items may not really be needed. Because like I said, if I were to keep really being more stringent about my reliability analysis I could end up with you know five items does anybody have any questions about that before we wrap up I mean obviously if you only have five items you can really only have like one factor or it would be probably pretty unusual to have more than one factor um because there's so few items that you're dealing with but um i guess it's a good thing that you know these items really kind of seem to hang together very well um when i look at this like like i said look at this if i was being really stringent and I hit and I trimmed it down to only five items I we actually get a chromex alpha 0.982 which is just kind of unheard of right so it's like there are some items there are some instruments like there's a depression um instrument that's only like four questions that's actually pretty So sometimes there are instruments that are very strong that only have a handful of items, but those are just, you know, they're not, they're not that common because usually you'd, you'd need to have, you know, more items in order to be able to assess what you're trying to assess. Trying to say something? Yeah, I do have a question, but it has to do with the APA formatting. Will, I mean, it might be better for, you know, individual, but I've done this before, straight copy and paste, and it was an APA format. Are we going to learn how to do APA format with this? well now when you say copy and paste you mean like you just took a table and you copied and pasted it well i mean yes i know that sometimes it'll say apa format for the spss and it'll come up like it's supposed to be apa format but i wouldn't know how to transfer the data to a paper So this, so I get, I'm just going to show you guys just for the last thing. This is this with the only five items. Of course, there's only one factor and it's, it's showing 94% of the variance. So that's kind of crazy, but interesting. Yes. For APA formatting, you, you shouldn't really be just like copying and pasting your tables or figures from SPSS or Jamovi. But I'm not, for this particular class, I'm not, you know, super strict about APA formatting because it's not really the focus of the class. You know, in in some other classes like rsm 801 you're really learning how to write up results and so that you know in that class it is more of an issue but you do you know i guess basically with tables for example you're only supposed to have horizontal lines that's apa style so you wouldn't have any vertical lines and that's the same thing in a figure you wouldn't have any you know vertical lines obviously you have your x and your y axis but you shouldn't have any lines over it so you know there's little things like that but really i'm you know if you just label your table and your figure like if you call it table one and you have a subheading and if you call it figure one and you have a subheading i would count that as being apa format because i don't expect you guys to adjust these um you know in order to be apa formatted for this particular assignment so i just want you to try to write it up in apa format but like if you were to change this to an apa format for example you'd have to take out all these lines i don't even like it's you know it's it's like it takes some extra you know elements like where you you wouldn't want to have any grid lines you know like you would be basically taking out these lines and things like that so it's it's just not necessary I think that would take too much time so um you know when I say when we say to try to write it up in APA style it's more just to try to write up your summary in apa style does that make sense put an example in the classroom too if anybody wants to see that but yeah you don't have to have your tables and formats to be perfectly apa formatted that's not really necessary for this assignment this is just a short assignment the main thing is just to kind of get a little experience with it and hopefully kind of part of what you guys are coming away from in this class is a sense of what it takes to design a good measure an instrument and to assess it for reliability and validity so what do you think do you think it's like easy to create a new measure or hard like is it like really quick or does it a lot of different steps you have to go through like any takeaways about what it takes to create new instruments i think definitely not quick or easy i mean i feel like it's an entire focus in and of itself mostly like if you're gonna get your doctorate in psychology or something and focus on psychometrics that's all you're really gonna do for a long period of time yeah yeah so hopefully you have the takeaway that which it sounds like you guys do, that for your dissertation, most likely, most everyone should be using validated, reliable instruments that are already out there, an existing instrument. Exactly. If you needed to, for some reason, create your own instrument, um it's probably best to try to adapt an existing one but that would just be a dissertation in itself whether you're creating your own instrument or adapting it or testing it with a special population like i know there's at least one student in the class where i said wow that actually would be really interesting for you to test that with your population because it is a unique and different population. But in most cases, like 99.9% of the time, use an existing instrument. And then you're able to look at, which is what I think most of you are more interested in is like relationships between variables or looking at whether there's group differences, that type of thing. Because if you're creating your own instrument, that's pretty much it. you'd just be doing that for your dissertation um joseph did you have a question or a comment yeah like adding just an item or two to measure something specific isn't creating a new instrument though right no no that's okay like during covid when i had students who were doing research during COVID, if I felt like it was relevant, we always had like some COVID questions. But I mean, I usually try, you know, there's actually a lot of research going on all the time, right? So it was actually like we found questions that you could, you know, you could either adapt or use. But like, let's say we're assessing something like stress. It may be that, you know, their stress could be impacted by not being able to go to school right if we were doing research with like i know one of my students was doing research with high school students and you know and they're looking at some things like physical activity and self -efficacy so maybe their physical activity was different it might have been more might have been less might have been the same you know so you can ask some questions like that and that's not creating a new instrument that's okay to add you know add some add some items or add some questions but um you know those just are kind of treated separately or you'd need to you know handle them differently but you can definitely do that on that that can be useful and interesting as well all right um well thank you guys for coming um if you have any other questions feel free to reach out to me. I'm happy to help out if you guys have any questions as we go through these last few weeks. After week eight, there is a one-week break. I know sometimes people don't know. So if you don't know, good news. There is a one-week break. There is a spring break. I think it's the first week in May. So there will be a break before your summer A class starts up so another professor told me that they're switching to blackboard ultra is that a different website at all or do you still just sign it in the same place um you still will just sign in the same place okay yeah so i don't think it will be too too different i mean we've all been going through training for it so i don't know i've gone through different a different place i guess like it's not organized the same way but yeah it's just yeah it's just a little different i think you'll like it i think it's you know there's some there's some cool new features with it but it's it's not anything that you need to worry about or that you need to you know log in differently or anything like that um but yeah and hopefully i'll see some of you guys at residency in july i don't know you know who's going to residency and who's not but um if you're going then um i look forward to seeing you guys there but in the meantime we do still have uh a little less than two weeks left of class so again feel free to reach out to me if you have any questions thank you all right thank you thank you dr kelly thank you bye Bye-bye. Thank you. Bye. Bye.