Yes, well, I just found myself watching your video for the third time because it was so helpful; your explanation addressed an issue that I (as a beginner) haven't dealt with yet. Sorry if I came across as snarky.
Hi, thanks for the video. I am combining many items of likert scale that measures one variable, in order to run regressions. The likert scale is from 1 to 5, strongly disagree to strongly agree. I am confused if I should use mean or sum to combute them? Cuz I thought the likert scale are ordinal data, so mean might not be appropriate. Any advice is highly appreciated. Thanks!
at 5:13 in the video simply name the variable something like "total" and then in the numerical expression box you can do either item1+item2+item3 OR you can write sum(item1,item2,item3). Either should work. Be careful using sum if you have missing data. Sum will sum whatever items have data and will ignore if other items have missing data. If you use the first approach and there is missing data on even just one item, you will have a missing value for the total score.
Hi! Great video! I have a question though, how can i combine the total scores of the subscales to get the total score as a whole and create a new variable?
Hi! great video. It is very helpful but I have some questions. 1. How do I analyse a likert scale data of one variable that have several items? I need to know which items give the most influence to the variable. (note that 1: strongly disagree 2:disagree 3: not sure 4: agree 5: strongly agree) 2. In your opinion, what is the best way to present the data? Is it by percentage, pie chart, mean core, etc? For example: What are the reasons for premarital screening? (mark according to scale) 1.1 to find out disease status 1.2 to avoid getting offspring with the disease 1.3 make future planning 1.4 consider to continue relationship thank you.
Hi! Thank you for this content this literally helps a lot. Aside from the presentation. How can I correlate a scale with another scale or variable considering that each scale contains a lot of questions?
You should combine the questions into a scale score using this video here and then see my video on correlation of two scale scores: ruclips.net/video/Kis6IwuqKjc/видео.html
Hi I have a question. How do i measure the Parental Authority Questionnaire using SPSS? I am confused as it does not depend on the total scores but there are subscales: Permissive/authoritarian etc. Please help me out if you can! Thank you.
Hi Marshyani, All you would need to do in this case is simply compute each subscale score the same way that I show you how to compute a total score. In other words, you simply have several "total" scores.
Hi. I'm trying to do a correlational study using 3 variables. Data for all three variables were collected using Likert scales. Therefore, I just wanted to know to convert the data from the Likert scale into a variable form so I can run a normality test and correlational test on it? In the video, at the computing part would I have to compute the mean scores of all the items or the median score or the sum?
Hi, how do you decide if you should combine items by either using the average or sum and also if a scale was not created to test a variable and you are using two different scales (eg. pay satisfaction and equity sensitivity) . How do you combine items of both scales to create one variable?
+J John this is a good question and the answer lies in whether you intended to measure a single underlying trait or construct with the group of items that you administered. Often I ask several questions all aimed at the same thing. Psychometric theory then allows me to average or sum the items to create a "measure" of a given construct. Often simple opinion polls or surveys don't include this approach, they assume a given item measure a given construct. So, it really comes down to how you planned the measurement approach. As for averaging or summing it is really a personal preference. I use average when there is missing data, but sometimes summed scores mean more to people familiar with the scales. Hope that helps.
Thanks for the video! Aside from what you mentioned about missing data in the comments below, is there a reason it is better to use the mean when calculating the score instead of a sum if you are sure there is no missing data?
Hi, you should be able to use the video and the scoring instructions you find on the web or in this article: Hamilton M. Development of a rating scale; for primary depressive illness. Br J Soc Clin Psychol. 1967 Dec;6(4):278-96. I don't know the HAM-D as well as other depression measures, but it should be pretty straightforward. That is, sum up the relevant items to obtain a total score. Creating categories is another matter, and you'll likely have to see the article or other articles to find the cutoff scores. To create cut off scores you would need to use "if, then" syntax like below: if(hamd=5 and hamd=10 and hamd=15)newvar=4. You will have to find the right scores to fill into the syntax above. Then paste this into a syntax window and execute the commands. I have a video showing how to run syntax too.
i just wanted to ask how i would use this in a linear regression, if i wanted to dummy code my variables for the regression when the mean totals brings up decimal places that are not represented by anything i guess would i then consider the scale as a continuous?
this should work fine. dummy coding will distinguish between say males and female and if the unstandardized coefficient is for instance .75 it means that moving the male (coded 0) to female (coded 1) on X corresponds to a .75 increase in the continuously measured dependent variable. so, females are higher than males. if the coefficient were -.75 then males would be higher than females.
so this method is available to give for example the variable of "attitude" a score based on 4 items ? and so to use it for the regression after or the test of KHI2 ?
How can I merge a few question (likert scale) to a histogram? for example question 1, 5, 10, dan 15 in item 1. So I want to combine all the questions become on single histogram.
well... I'm not sure I entirely understand the question... You can create summed scores only if the items are meant to measure the same thing. Otherwise, it would be best to keep them separate.
I have a data of different firms I have to first identify whether they are bankrupt or not then I have inventory values and cost öf goods sold from them I have to calculate inventory turnover but these data are given for different firms and each firms data is for 6 to 7 years having 4 quarter, inventory turnover will be calculated for each firm but on yearly basis
Sorry, I don't quite fully understand your question... I think "merging" as you mention is simply the same as summing up items in this example. So you could use the same process here to "merge" all the data into a single turnover variable.
Hi. Is there any ways that I could contact you? I have several questions regarding my study and i'm not to sure on how to score the questionnaire. Looking forward to your reply. Thank you.
Thanks for the video, but it finishes short! After we have calculated the mean scores at the end, what do we do and how do we interpret those results? Thanks
Hi Tom, I don't mean to be trite, but you can do whatever you wish with the mean scores. In short, after computing them, you can now do t-tests or correlations, etc. The computation of the scores themselves is usually only part of a larger hypothesis that will require further analysis.
Sorry if I came across wrong, I'm just not sure what to do with it afterwards. I am trying to correlate 2 variables together to see if as one increases the other does too. Would it be ok to use the mean scores and treat them as continuos data then to use in partial correlations etc? Thanks
No worries Tom. I just didn't want to sound too abrupt in my reply. Yes, of course, what you describe is exactly the kind of thing that would be useful. Use the mean scores as the variables in the correlations rather than the individual items scores (would be too many to do). That's exactly why you compute the summary mean scores.
Ah I understand now. Just one final question if thats ok. My problem is that by using mean scores I now have some means that are for example 4.5. But, 4.5 isn't a possible answer on the original likert scale so how do I interpret these new numbers? For example, in my findings discussion surely I can't say "20% of people scored between agree and strongly agree" can I? Hope you understand :)
Yes, decimal values will be not only possible, but likely. 4.5 is the average of responses from a person who gave mostly 4s and 5s on the Likert items. You can give the percentages of people with AVERAGE of 4 or greater, 4 or less, etc. as you normally would.
hello, thanks very much for this video. pls how can I merge strongly agree and agree to have agree and strongly disagree and disagree to have disagreements to determine certain percentage of attitude.
the first part of this video should help: ruclips.net/video/8YoiEdFZ9GY/видео.html. What you want to do is recode values in SPSS and in your case you'll probably recode 1(SD) and 2(D) into 1 and 3(A) and 4(SA) into 2. That will give you just two values where 1 is strongly disagree or disagree and 2 is agree or strongly agree. I hope that helps.
hello, thks for this video. i have a question: if the scale is multidimensionnal, does it make sens to calculate a score for the whole scale? if it is, how to compute it? is it the average of all items in all dimentions?
In many cases it does make sense to compute a total score and you would do it just as you have described. Sometimes scales don't have total scores and it is always best to refer to the original article where the scale was first published or refer to the test manual if it is available. Hope this helps.
Loren L. Toussaint thank you so much. Few years ago, I started to write a paprer (that I gave up) and I had to deal with something like that. When I performed PCA analysis for trust scale it always showed 3 dimensions (credibility, integrity and goodwill) and I asked myself if it was right to compute a total score of the (3×4) items and talk about trust ...
Mariah Sandrine Dating you want to compute variables to follow standard measurement practices in psychology and other social sciences. Averaging or summing items from a scale reduces measurement error and allows you a better, more accurate, assessment of the true construct you are trying to study.
After the mean of a scale is computed you can compare it to other means of other scales to see which is higher/lower. For instance, a 10 item depression scale has an average of 7 and a 10 item anxiety scale has an average of 3. Hence, anxiety symptoms are lower than depression. In other words, every person's items get averaged into a scale score (sum or mean) and then these scale scores are averaged for the group.
Good video. Please how can you can the regression using these data. You grouped them into three so can you please show me how to run regression please?
Hi, There are two videos that I have posted on my channel that show you how to run simple regression as well as multiple regression Two variable Regression is here: ruclips.net/video/JQBMbRkp7WY/видео.html Multiple Regression is here: ruclips.net/video/L9yFGtscyL8/видео.html
Thanks a lot. What i meant is that, how will i make those variables to become one main variable? I have a 3 main variables. Training and compensation are independent variables and performance as dependent variable. In training, I have 29 statements or questions using likert scale, compensation 20 and performance 24. So if i enter those data into SPSS, it will be like 73 variables altogether. So how will i put data for training under main variable training, compensation and performance? Will i use the transform and add all the variables for training to be in one column and compensation same and performance as well?
Ok, so you want to use the 3 variables that you compute using the video above in your regression analyses. You take the 29 statements in the training variable and average or sum them as shown in the video and then use that computed variable in further analyses. Its the compute variables that are of interest, not the items. Hope that helps.
Hi, I think you're after a t-test or correlation. t-test is here:ruclips.net/video/m8gXTYoBMYw/видео.html, but you can go to my channel and find other t-tests and correlation videos.
Hi what do I do if I ran the reliability test but it displays the message ‘the value is negative due to a negative average covariance among items. This violates reliability model assumptions. You may want to check item codings’ even if I have reverse coded? I am using the Rosenberg self esteem scale please help I’m so bad with stats! :(
Hi Zay, Even if you believe you have recoded correctly, it is possible that due to some other factor the scale is not reliable. usually it is an item or two. I am familiar with the RSE, but I'm not sure why this is happening. However, there are difficulties with reversed items when English scales are used in other cultures.
Loren L. Toussaint I have recoded the correct variables and it still isn’t working, are my items labelled wrong? My values are 1= strongly disagree 2=Disagree 3=Agree 4= strongly agree I think recoded these for some of the questions as stated by the RSE. I am entering the values as items is that incorrect?
@@hanik8918 sounds like everything is correct. Try getting alpha for just positive worded items (not reversed). Where (what country and language) are you using the scale?
@@loren.toussaint Is there a benefit to average over sum or sum over average? For example if I want to check if there is a correlation between a persons occupation and a 10 item questionnaire?
@@SkidMcmarxx generally no difference. However, if there is a small amount of missing data, you might save some cases from being deleted by computing the mean (which is unaffected by the missing data). For instance, a person who answers 9 of 10 questions will have a lower sum score because they have one less item to add to the sum, but if you average the items the average is unaffected by having 9 or 10 responses. If you choose mean, you'll have to choose a cutoff where too much missing data means you can't compute the mean. Often that is determined to be 50% of the items, so anyone with 5 or more responses, you compute the mean, anything less and it is missing. I hope that helps.
i have to find a variable for a survey.i have to find the number of males and females who voted for a certain question with best(1) and second best (2) to generate a satisfactory survey.how to compute a variable like this ? where a variable v2200_1 has two values one and two and plot them in a crosstab against gender..
Hi Trishita, I'm sorry. I don't entirely understand your question, but I think you're asking about how to run a crosstab in SPSS. Is that correct? If so, its under the "analyze," "descriptive" menu. Simply put each variable into the row or column. Click the statistics button and choose "chi-square." Then click on "cells" and choose "row" or "column." You don't need both. Hope that helps.
i know how to make a crosstab ..but i want a collective row variable for option one and two against the column variable.Let me explain you a scenario..A question can be ranked with 4 ratings, 1,2,3,4.Suppose the variable for that is V20 which can have any of these valuess.My task is to form a crosstab which will show the satisfaction index only for rating 1 and 2 against the genders Male and Female.That means I have to find the value of those men and women who support rating 1 and 2 and create a new variable which denotes both the ratings(V20=1 | V20 =2).But my result is displaying separate counts for rating 1 and rating 2 not the overall
Ok, I think I got it now. You want to use the recode procedure only that runs in this video from about 1:50 to 3:20. Then analyze the recoded variable, don't follow any more of the video after that.
Yes, you want to recode as follows. SPSS syntax command is: recode v20 (1=1)(2=1)(3=2)(4=2)into v20recoded. note: i've recoded 1 and 2 into 1 and 3 and 4 into 2 you can also accomplish this in the recode menu as seen on the video. Good luck!
Hi loren. Do we have a cutt off score that differentiates between a high/low score? I have an instrument with 6 items and a 5 point likert scale. How to i determine a cutt off score? or if its continuous, as in the person is high/ low on a given construct, is there any specific point (number) to differentiate them?
@@loren.toussaint i was wondering what you would do if the scales are different so for example one is on a 5 point scale while one is at a 7 point scale, this is for a uni work so im sure they added that on purpose
@@laaaaaaaa533 I'd probably standardize the items first, then compute a scale score. There are other methods, but standardization is meant exactly for this purpose--to make two different response scales equal.
this video shows how to do a simple composite. if you need to weight items then you have to do that first and the compute the composite. weight scores would give more weight or impact to one item as compared to another. for example, (item 1 * 2) + (item 2) + (item 3 * 4) or something like that. item 1 gets twice the weight and item 3 gets four times the weight of item 2 which just gets a regular weight of 1.
+Loren L. Toussaint Thank you for your quick response. I have two likert scales that I need to run correlation test on. I was wondering if you have a video on this? i am really confused :(
+Loren L. Toussaint thanks again. I've watched them, though it doesn't explain how to check correlation between two different likert scales. I know how to run correlation between similar likert scales but not on different ones. What I'm really trying to do is validate my scale against an already validated scale by using correlation but my scale is 7 point while the validated scale is 6 point. Would you know how to do this on spss? Thanks again.
+BravoEchoAlpha You're good to go if you know how to run Pearson correlation. Pearson standardizes the variables so it doesn't matter the scales the variables are on. For instance, people correlate GPA (4 point scale) with SAT (hundreds of points) all the time. Scales don't matter for correlation.
+Rebekahgzls Hi, thanks for your comment. Many questionnaires have reverse-keyed items and you MUST reverse score them before computing the total score or the computed total will be meaningless. If you have a questionnaire that does not have reverse keyed items (be sure to double check this), then you can skip that step.
Hello there. I want to know that after running analysis i have an output sheet in which i have analysis report. Now i want to know that which values from output i have to put in results section concerned to item total corelation of scale. Like i cant understand......urgent help. Please
generally all you need to include is the coefficient alpha value itself and typically alphas of at least .70 are expected. you typically woudn't include the item-total correlations in a formal paper or report. hope that helps.
u saved my life. thank u so much for uploading this video its super clear and easy to follow. i was in tears trying to figure this out. god bless.
so glad it was helpful!
This helped me so much finding an average scale positive for my applied stats class! Thank you so much for doing these videos!
So glad to hear it! Thanks for your kinds words!!
Yes, well, I just found myself watching your video for the third time because it was so helpful; your explanation addressed an issue that I (as a beginner) haven't dealt with yet. Sorry if I came across as snarky.
No snark taken :-) Just wanted to be sure you understood that you couldn't skip that step with many questionnaires.
Hi, thanks for the video. I am combining many items of likert scale that measures one variable, in order to run regressions. The likert scale is from 1 to 5, strongly disagree to strongly agree. I am confused if I should use mean or sum to combute them? Cuz I thought the likert scale are ordinal data, so mean might not be appropriate. Any advice is highly appreciated. Thanks!
Taking a 'Median' is much better instead of 'Mean' while computing a variable...
Thanks! Very lucidly explained.
Thank you! :D This was a great help for computing scores for a quiz I am doing for my undergraduate dissertation.
Awesome! Glad you found it useful!!
Ah, excellent video. Very good teaching! Thank you!
Thank you 🙏🏾
Hello, thank you for a great tutorial. If I want to compute the total score for each respondent, instead of total mean?
at 5:13 in the video simply name the variable something like "total" and then in the numerical expression box you can do either item1+item2+item3 OR you can write sum(item1,item2,item3). Either should work. Be careful using sum if you have missing data. Sum will sum whatever items have data and will ignore if other items have missing data. If you use the first approach and there is missing data on even just one item, you will have a missing value for the total score.
Thanks a lot for this video.
You have been very helpful!
Hi! Great video! I have a question though, how can i combine the total scores of the subscales to get the total score as a whole and create a new variable?
Typically you would sum or average all items and not divide them by subscales. Hope that helps.
Hi, thank you so much for been very helpful.
Hi! great video. It is very helpful but I have some questions.
1. How do I analyse a likert scale data of one variable that have several items? I need to know which items give the most influence to the variable. (note that 1: strongly disagree 2:disagree 3: not sure 4: agree 5: strongly agree)
2. In your opinion, what is the best way to present the data? Is it by percentage, pie chart, mean core, etc?
For example:
What are the reasons for premarital screening? (mark according to scale)
1.1 to find out disease status
1.2 to avoid getting offspring with the disease
1.3 make future planning
1.4 consider to continue relationship
thank you.
A pie chart would be just fine in this case, sometimes bar charts work too, but it is totally your preference. They both convey the same meaning.
Hi! Thank you for this content this literally helps a lot. Aside from the presentation. How can I correlate a scale with another scale or variable considering that each scale contains a lot of questions?
You should combine the questions into a scale score using this video here and then see my video on correlation of two scale scores: ruclips.net/video/Kis6IwuqKjc/видео.html
Hi I have a question. How do i measure the Parental Authority Questionnaire using SPSS? I am confused as it does not depend on the total scores but there are subscales: Permissive/authoritarian etc. Please help me out if you can! Thank you.
Hi Marshyani, All you would need to do in this case is simply compute each subscale score the same way that I show you how to compute a total score. In other words, you simply have several "total" scores.
Hi. I'm trying to do a correlational study using 3 variables. Data for all three variables were collected using Likert scales. Therefore, I just wanted to know to convert the data from the Likert scale into a variable form so I can run a normality test and correlational test on it? In the video, at the computing part would I have to compute the mean scores of all the items or the median score or the sum?
typically you would compute the mean of the items to arrive at a variable score
Do we use this mean column to go for further analysis like correlation ?
Yes, that is correct.
Hi, how do you decide if you should combine items by either using the average or sum and also if a scale was not created to test a variable and you are using two different scales (eg. pay satisfaction and equity sensitivity) . How do you combine items of both scales to create one variable?
+J John this is a good question and the answer lies in whether you intended to measure a single underlying trait or construct with the group of items that you administered. Often I ask several questions all aimed at the same thing. Psychometric theory then allows me to average or sum the items to create a "measure" of a given construct. Often simple opinion polls or surveys don't include this approach, they assume a given item measure a given construct. So, it really comes down to how you planned the measurement approach. As for averaging or summing it is really a personal preference. I use average when there is missing data, but sometimes summed scores mean more to people familiar with the scales. Hope that helps.
Thanks for the video! Aside from what you mentioned about missing data in the comments below, is there a reason it is better to use the mean when calculating the score instead of a sum if you are sure there is no missing data?
If there are no missing data, then mean and sum will be equally useful. No advantage to one or the other.
@@loren.toussaint thanks!
I want to calculate the Hmailtion depression scale 24 variables and then to categorize it to four categories ???????
Hi, you should be able to use the video and the scoring instructions you find on the web or in this article: Hamilton M. Development of a rating scale; for primary depressive illness. Br J Soc Clin Psychol. 1967 Dec;6(4):278-96. I don't know the HAM-D as well as other depression measures, but it should be pretty straightforward. That is, sum up the relevant items to obtain a total score. Creating categories is another matter, and you'll likely have to see the article or other articles to find the cutoff scores. To create cut off scores you would need to use "if, then" syntax like below:
if(hamd=5 and hamd=10 and hamd=15)newvar=4.
You will have to find the right scores to fill into the syntax above. Then paste this into a syntax window and execute the commands. I have a video showing how to run syntax too.
i just wanted to ask how i would use this in a linear regression, if i wanted to dummy code my variables for the regression when the mean totals brings up decimal places that are not represented by anything i guess would i then consider the scale as a continuous?
this should work fine. dummy coding will distinguish between say males and female and if the unstandardized coefficient is for instance .75 it means that moving the male (coded 0) to female (coded 1) on X corresponds to a .75 increase in the continuously measured dependent variable. so, females are higher than males. if the coefficient were -.75 then males would be higher than females.
so this method is available to give for example the variable of "attitude" a score based on 4 items ? and so to use it for the regression after or the test of KHI2 ?
correct. not sure what you're referring to regarding KHI2, but your understanding appears to be correct.
How can I merge a few question (likert scale) to a histogram?
for example question 1, 5, 10, dan 15 in item 1. So I want to combine all the questions become on single histogram.
well... I'm not sure I entirely understand the question... You can create summed scores only if the items are meant to measure the same thing. Otherwise, it would be best to keep them separate.
I have a data of different firms I have to first identify whether they are bankrupt or not then I have inventory values and cost öf goods sold from them I have to calculate inventory turnover but these data are given for different firms and each firms data is for 6 to 7 years having 4 quarter, inventory turnover will be calculated for each firm but on yearly basis
Yes, you could do this if you treated each data time point like an item in my example
Hi I need to calculate the cumulative percentage of responses, with the lickert scale, how do i do it?
you need to look at the cumulative percentage column in the frequencies analysis
thanks for sharing, the video was helpful alot. 1'm running a similar analysis, i hope i can consult whenever i want? thanks again.
hope to help as much as I can. :-)
hi, I want to ask what if we want to calculate inventory turnover but för the quarterly data and that may require the merging??
Regards
Sorry, I don't quite fully understand your question... I think "merging" as you mention is simply the same as summing up items in this example. So you could use the same process here to "merge" all the data into a single turnover variable.
Hello, thanks for this video, can I use the mean scores as datas for computing ANOVA?
Yes, this is exactly how you should use these scores.
@@loren.toussaint thank you!
Hi. Is there any ways that I could contact you? I have several questions regarding my study and i'm not to sure on how to score the questionnaire. Looking forward to your reply. Thank you.
you'll find my email on www.luther.edu/touslo01. you can email me directly if you like. you can also find me on ResearchGate.
Thanks for the video, but it finishes short! After we have calculated the mean scores at the end, what do we do and how do we interpret those results? Thanks
Hi Tom, I don't mean to be trite, but you can do whatever you wish with the mean scores. In short, after computing them, you can now do t-tests or correlations, etc. The computation of the scores themselves is usually only part of a larger hypothesis that will require further analysis.
Sorry if I came across wrong, I'm just not sure what to do with it afterwards. I am trying to correlate 2 variables together to see if as one increases the other does too.
Would it be ok to use the mean scores and treat them as continuos data then to use in partial correlations etc?
Thanks
No worries Tom. I just didn't want to sound too abrupt in my reply. Yes, of course, what you describe is exactly the kind of thing that would be useful. Use the mean scores as the variables in the correlations rather than the individual items scores (would be too many to do). That's exactly why you compute the summary mean scores.
Ah I understand now. Just one final question if thats ok.
My problem is that by using mean scores I now have some means that are for example 4.5. But, 4.5 isn't a possible answer on the original likert scale so how do I interpret these new numbers?
For example, in my findings discussion surely I can't say "20% of people scored between agree and strongly agree" can I?
Hope you understand :)
Yes, decimal values will be not only possible, but likely. 4.5 is the average of responses from a person who gave mostly 4s and 5s on the Likert items. You can give the percentages of people with AVERAGE of 4 or greater, 4 or less, etc. as you normally would.
hello, thanks very much for this video. pls how can I merge strongly agree and agree to have agree and strongly disagree and disagree to have disagreements to determine certain percentage of attitude.
the first part of this video should help: ruclips.net/video/8YoiEdFZ9GY/видео.html. What you want to do is recode values in SPSS and in your case you'll probably recode 1(SD) and 2(D) into 1 and 3(A) and 4(SA) into 2. That will give you just two values where 1 is strongly disagree or disagree and 2 is agree or strongly agree. I hope that helps.
hello, thks for this video. i have a question: if the scale is multidimensionnal, does it make sens to calculate a score for the whole scale? if it is, how to compute it? is it the average of all items in all dimentions?
In many cases it does make sense to compute a total score and you would do it just as you have described. Sometimes scales don't have total scores and it is always best to refer to the original article where the scale was first published or refer to the test manual if it is available. Hope this helps.
Loren L. Toussaint thank you so much.
Few years ago, I started to write a paprer (that I gave up) and I had to deal with something like that. When I performed PCA analysis for trust scale it always showed 3 dimensions (credibility, integrity and goodwill) and I asked myself if it was right to compute a total score of the (3×4) items and talk about trust ...
I think that would be just fine. Its common to do so.
what's the purpose of computing for the variable of a scale?
Mariah Sandrine Dating you want to compute variables to follow standard measurement practices in psychology and other social sciences. Averaging or summing items from a scale reduces measurement error and allows you a better, more accurate, assessment of the true construct you are trying to study.
thanks! :)
So after this mean is computed, how can I go on and compute 1 value for the scale so that it can be compared to another likert scale?
After the mean of a scale is computed you can compare it to other means of other scales to see which is higher/lower. For instance, a 10 item depression scale has an average of 7 and a 10 item anxiety scale has an average of 3. Hence, anxiety symptoms are lower than depression. In other words, every person's items get averaged into a scale score (sum or mean) and then these scale scores are averaged for the group.
Good video. Please how can you can the regression using these data. You grouped them into three so can you please show me how to run regression please?
Hi,
There are two videos that I have posted on my channel that show you how to run simple regression as well as multiple regression
Two variable Regression is here: ruclips.net/video/JQBMbRkp7WY/видео.html
Multiple Regression is here: ruclips.net/video/L9yFGtscyL8/видео.html
Thanks a lot. What i meant is that, how will i make those variables to become one main variable? I have a 3 main variables. Training and compensation are independent variables and performance as dependent variable. In training, I have 29 statements or questions using likert scale, compensation 20 and performance 24. So if i enter those data into SPSS, it will be like 73 variables altogether. So how will i put data for training under main variable training, compensation and performance? Will i use the transform and add all the variables for training to be in one column and compensation same and performance as well?
Ok, so you want to use the 3 variables that you compute using the video above in your regression analyses. You take the 29 statements in the training variable and average or sum them as shown in the video and then use that computed variable in further analyses. Its the compute variables that are of interest, not the items. Hope that helps.
Thank you very much. Now I get it. Thank you very much.
can you please send me the link of the next video, that how you computed these questions when they are in decimal places...
Hi Shahrukh, Can you please say more about what you need? I'm not sure what you're asking... Decimal values are perfectly acceptable in this process.
Hello Loren,
So for example when If I take sample mean 3 and get a value for mlq as in all mlq questions 3.7 then will it be a significant study?
Hi, I think you're after a t-test or correlation. t-test is here:ruclips.net/video/m8gXTYoBMYw/видео.html, but you can go to my channel and find other t-tests and correlation videos.
Thanks a lot😊
Thanks!!
Hi what do I do if I ran the reliability test but it displays the message ‘the value is negative due to a negative average covariance among items. This violates reliability model assumptions. You may want to check item codings’ even if I have reverse coded? I am using the Rosenberg self esteem scale please help I’m so bad with stats! :(
Hi Zay, Even if you believe you have recoded correctly, it is possible that due to some other factor the scale is not reliable. usually it is an item or two. I am familiar with the RSE, but I'm not sure why this is happening. However, there are difficulties with reversed items when English scales are used in other cultures.
Loren L. Toussaint hi thanks for your reply! How can I tell which items to get rid of?
Loren L. Toussaint I have recoded the correct variables and it still isn’t working, are my items labelled wrong? My values are 1= strongly disagree 2=Disagree 3=Agree 4= strongly agree I think recoded these for some of the questions as stated by the RSE. I am entering the values as items is that incorrect?
@@hanik8918 sounds like everything is correct. Try getting alpha for just positive worded items (not reversed). Where (what country and language) are you using the scale?
Loren L. Toussaint England, United Kingdom!
Why do you compute the average?
you can compute average or sum. its just the way to create a scale score from individual items.
@@loren.toussaint Is there a benefit to average over sum or sum over average? For example if I want to check if there is a correlation between a persons occupation and a 10 item questionnaire?
@@SkidMcmarxx generally no difference. However, if there is a small amount of missing data, you might save some cases from being deleted by computing the mean (which is unaffected by the missing data). For instance, a person who answers 9 of 10 questions will have a lower sum score because they have one less item to add to the sum, but if you average the items the average is unaffected by having 9 or 10 responses. If you choose mean, you'll have to choose a cutoff where too much missing data means you can't compute the mean. Often that is determined to be 50% of the items, so anyone with 5 or more responses, you compute the mean, anything less and it is missing. I hope that helps.
@@loren.toussaint Yes, thank you so much.
bless u
i have to find a variable for a survey.i have to find the number of males and females who voted for a certain question with best(1) and second best (2) to generate a satisfactory survey.how to compute a variable like this ? where a variable v2200_1 has two values one and two and plot them in a crosstab against gender..
Hi Trishita, I'm sorry. I don't entirely understand your question, but I think you're asking about how to run a crosstab in SPSS. Is that correct? If so, its under the "analyze," "descriptive" menu. Simply put each variable into the row or column. Click the statistics button and choose "chi-square." Then click on "cells" and choose "row" or "column." You don't need both. Hope that helps.
i know how to make a crosstab ..but i want a collective row variable for option one and two against the column variable.Let me explain you a scenario..A question can be ranked with 4 ratings, 1,2,3,4.Suppose the variable for that is V20 which can have any of these valuess.My task is to form a crosstab which will show the satisfaction index only for rating 1 and 2 against the genders Male and Female.That means I have to find the value of those men and women who support rating 1 and 2 and create a new variable which denotes both the ratings(V20=1 | V20 =2).But my result is displaying separate counts for rating 1 and rating 2 not the overall
Ok, I think I got it now. You want to use the recode procedure only that runs in this video from about 1:50 to 3:20. Then analyze the recoded variable, don't follow any more of the video after that.
reverse them? but i dont want to recode 1 to anythin..i want to join 1 and 2 to have the same value
Yes, you want to recode as follows. SPSS syntax command is:
recode v20 (1=1)(2=1)(3=2)(4=2)into v20recoded.
note: i've recoded 1 and 2 into 1 and 3 and 4 into 2
you can also accomplish this in the recode menu as seen on the video.
Good luck!
what does the mean of an item represent?
its the same as a total score on a test. it represents if the person is high/low on given construct.
Hi loren. Do we have a cutt off score that differentiates between a high/low score? I have an instrument with 6 items and a 5 point likert scale. How to i determine a cutt off score? or if its continuous, as in the person is high/ low on a given construct, is there any specific point (number) to differentiate them?
thank youuu!!!!!!!!
You are welcome! Glad it was useful!!
@@loren.toussaint i was wondering what you would do if the scales are different so for example one is on a 5 point scale while one is at a 7 point scale, this is for a uni work so im sure they added that on purpose
@@laaaaaaaa533 I'd probably standardize the items first, then compute a scale score. There are other methods, but standardization is meant exactly for this purpose--to make two different response scales equal.
@@laaaaaaaa533 yes. Standardize items, then compute scale scores, then do regression analyses. In that order.
@@loren.toussaint thank you soo much really grateful
How to calculate composite score for likert scale
Yes, this video shows you how to do this for a Likert scale. Just follow the procedure. Hope that is helpful.
what is the difference between composite score and weighted score
Thanks
this video shows how to do a simple composite. if you need to weight items then you have to do that first and the compute the composite. weight scores would give more weight or impact to one item as compared to another. for example, (item 1 * 2) + (item 2) + (item 3 * 4) or something like that. item 1 gets twice the weight and item 3 gets four times the weight of item 2 which just gets a regular weight of 1.
what if the scales are different? e.g 7 scale VS 6 scale? thank you! :)
Often folks standardize the variables and then compute the scale score when scales are different
+Loren L. Toussaint Thank you for your quick response. I have two likert scales that I need to run correlation test on. I was wondering if you have a video on this? i am really confused :(
I have vids on Pearson correlation, linear regression, and multiple correlation and regression that would probably be helpful...start with Pearson vid
+Loren L. Toussaint thanks again. I've watched them, though it doesn't explain how to check correlation between two different likert scales. I know how to run correlation between similar likert scales but not on different ones. What I'm really trying to do is validate my scale against an already validated scale by using correlation but my scale is 7 point while the validated scale is 6 point. Would you know how to do this on spss? Thanks again.
+BravoEchoAlpha You're good to go if you know how to run Pearson correlation. Pearson standardizes the variables so it doesn't matter the scales the variables are on. For instance, people correlate GPA (4 point scale) with SAT (hundreds of points) all the time. Scales don't matter for correlation.
The first part on recoding the variable 9 should have been left out. Confusing and irrelevant for us beginners.
+Rebekahgzls Hi, thanks for your comment. Many questionnaires have reverse-keyed items and you MUST reverse score them before computing the total score or the computed total will be meaningless. If you have a questionnaire that does not have reverse keyed items (be sure to double check this), then you can skip that step.
Hello there. I want to know that after running analysis i have an output sheet in which i have analysis report. Now i want to know that which values from output i have to put in results section concerned to item total corelation of scale. Like i cant understand......urgent help. Please
generally all you need to include is the coefficient alpha value itself and typically alphas of at least .70 are expected. you typically woudn't include the item-total correlations in a formal paper or report. hope that helps.