Thank you so much for this very simple yet very precise explanation. This will help me test the reliability of my Thesis. At least I can solve it on my own before letting my Statistician check my work. God bless and thank you for sharing this.
ARGH!!!! Thank you so much for this video! I can’t say I love stats but I now understand this cronbach’s alpha a little more and hence LOVE this video. Gracias!
This is an excellent video- each step is demonstrated in the process of arriving at Cronbachs alpha. This is the first steo in the quantitative/qualitative results for my research. Thank you
Thank you so much!!! We have to report this for our topic "reliability of research instruments" and it is so difficult to understand :))) New subscriber here ♡
I am now confident to perform the Cronbach's Alpha after watching your lecture, sir Glen. The discussion is short but sweet. Thank you so much po, I'm your fellow faculty member in UEP sir, Ms. Abenes po ito. Thank yo pooo
You are very much welcome madam! Here is an actual research paper questionnaire by way: ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-H8Tlxu9XvEE.html
Thank you for posting this video, extremely helpful! Btw, I wonder if I have more than one section in my questionnaire, do I need to get the Cronbach alpha for each section separately? Hope you can answer this because it helps me in completing my research, thanks again.
Hi, thanks for asking. If the sections are measuring different concepts then you will need to do separate Cronbach's alphas. If the sections are measuring the same concepts like disagree/agree and unlikely/likely have similar meanings in relationship to this concept, then you only need to do one Cronbach’s alpha. You can check out more frequently asked question in this site: www.real-statistics.com/reliability/internal-consistency-reliability/cronbachs-alpha/
Excellent! Clear and succinct tutorial. For the time I now understand Kronbach's Alpha! I've subscribed. Looking forward to seeing all you videos. Keep 😊🌺
Thank you so much for your great videos, I can easily understand your explanation. Do you have a video talking about checking/testing the validity of a questionnaire?
Hi, I also asked this to myself. Though I tried it both and I am not sure if the difference would be significant. I am till reading some articles on this...
Thank you sir for the clear explanation, may I know if you have a reference for the interpretation of the alpha level. Thank you in advance sir for any response
Hi, thanks for asking. The result should be certified by a statistician or your statistic professor. Sometimes statisticians would also recompute and suggest removing some items to increase the reliability of your survey questionnaire.
Does the scale interpretation matter in measuring the cronbach's alpha? Would the result change if 1 is the highest and 5 being the lowest or vice versa?
During the pilot test conducted on 20-50 respondents, should we focus solely on checking reliability, or is it necessary to conduct factor analysis at this stage? Please clarify the order of priority and what is mandatory during the pilot testing stage.
I have 30 items do i need to include all 30 items for calculating the Cronbach Alpha just like you did or i need to split the items based on my three research questions?
Hi, thanks for asking. You can include all 30 items for your alpha calculation. However, if you want to check more on the internal consistency of the group of items based on your research questions, you may split them to check if the answers are highly correlated.
Hi, thanks for asking. You can apply KR20 for your nominal dichotomous measures (with binary coding, say 1-yes and 0-no) and Cronbach's alpha for the Likert scale questions. But even if you apply/use Cronbach's alpha on dichotomous measures, you will notice it yields the same result because the formula is just the same. Here is an example: ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-FuBRfWzyaZo.html
Hi, thanks for asking. Taking the variance was shown in the video using the MS excel formula. But if you would like to solve it manually, you can follow the video in this link: ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-IRLywOyWzB8.html
This video is very informative! Very easy to follow. But I was just wondering if what would happen if I get an alpha below 0.70, will that affect my sample size in the study?
Hi, thanks for asking. If your alpha is below 0.70, then it has detected some reliability issues that require you to modify the survey or questionnaire. Cronbach's alpha result does not affect the sample size of the study, rather its like a go signal that the survey questionnaire is ready and approved for floating when it passed the reliability test.
@@eirenevizconde1065 Hi, thanks for asking. You may use the Yamane's formula or normally called slovin's formula: You can check out this example: ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-0c6d30-tjks.html
in 7 point likert scale and 5 point likert scale same cronbach formula applicable? plzz do inform...eagerly waiting your reply....cz i m stuck in this question...
Hi, thanks for asking. Theoretically, Cronbach’s alpha results should give you a number from 0 to 1, but you can also get negative numbers. A negative number indicates that something is wrong with your data-it is possible that some score items were reversed...
Thank You for an excellent explanation. I want to ask you that My questionnaire includes both dichotomous (yes/no) and multiple choice responses. In this case, do I need to separate dichotomous and multiple responses from one question and then apply RK20 reliability test? I would be very grateful to you. Thank you!!
Hi, thanks for asking. Cronbach's alpha is intended for Likert scale survey questionnaire. If you are trying to test the reliability of a multiple choice test, then you should use KUDER-RICHARDSON. Here is an example: ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-FuBRfWzyaZo.html
Hello! Sir.. If a questionnaire contains dichotomous, and other mix types of questions which have more than 2 options... Is it possible to check reliability and validity of that questionnare?
Hello Sir! Is this only possible for likert scales? Or is it okay to be used for multiple choice survey questionnaire? Thank you btw for your video Sir, it is very well explained and easy to understand.
Hi, Thanks for asking. For multiple choice questionnaire, you may want to try Kuder-Richardson Formula 20 (KR20). Here is a sample video: ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-FuBRfWzyaZo.html
@@glentox1563 Thank you Sir for your response 💕 I just finished watching the video you recommended, I really understood it well. However I'm confused because my MCQ questionnaire has no correct answers because I'm trying to find out their behavior, it's subjective in a way. It's like similar to a checklist Sir. Is it okay to still use the KR20 formula for this questionnaire?
Hi, thanks for your query. Basically, KR20 and Cronbach's alpha are the same. Cronbach's alpha is the extension of KR20. KR20 works only for dichotomous measures while the other works with polytomous measures. You can apply KR20 for your nominal dichotomous measures (with binary coding, say for example, 1-yes and 0-no) and Cronbach's alpha for the likert scale questions. But even if you apply/use Cronbach's alpha on dichotomous measures, you will notice it yields same result because the formula is just the same.
