Cronbach's Alpha (Simply explained)
Вставка
- Опубліковано 28 вер 2024
- Cronbach's alpha (or tau-equivalent reliability) is a measure of the relationship between a group of questions. The group of questions is called a scale and each question in that group is an item. Cronbach's alpha is therefore a measure of the internal consistency of a scale and thus of the strength of its reliability.
More details about Cronbach's alpha
datatab.net/tu...
And the Cronbach's alpha calculator
datatab.net/st...
If you like, please find our e-Book here: datatab.net/statistics-book 😎
This is the only video that has satisfied my curiosity about Cronbach’s alpha. Thank you so much!
Many thanks for your nice feedback! Regards Hannah
I love how you show us the examples. Very clear and practical!
Glad it was helpful!
Thank you so much for your video! your explanation is so intuitive and so clear
Many thanks for the Feedback! Regards, Hannah
thank you for the best and crisp explanation. it does not take so much time yet precise. it would be very useful if you teach Multi level modelling and structural equation modelling using AMOS
Very clear and simple explanation!
Many thanks! 😊Regards, Hannah
thank you for your short and precise explanation.
Thank you for being clear, concise and informative. The video was very helpful!
Glad it was helpful!
Thank you for your contribution! I created a proficiency test which assesses English idiom knowledge. Questions are only multiple choice type. Is it enough to measure cronbach's alpha in this case?
Can this be used with non-latent variables, actual scientific measurements, to determine if there is reliability in multiple trials of a scientific experiment and if more trials are needed to achieve data reliability?
so good and easy to understand
Many thanks!
Great video, thank you!
Glad you liked it!
thanks for the video , very helpful video.
Thank you! Whats the difference between Cronbach's alpha and ICC?
Hmm, I think Cronbach's alpha is a special case of ICC, but I have not investigated this and therefore consider the answer with caution! Regards Hannah
How do I handle the test if I have missing values? In case I have an option of NA.
Thank you for the video!
Thank you so much
Hi there, when you say "the goal now is that the responses to the various items match well, i.e., are highly correlated", does it matter if some questions are highly positively correlated while other questions are highly negatively correlated? How does the sign of the correlation affect Cronbach's alpha?
Im pretty sure it doesnt matter if they are positively or negatively correlated. The aim is to measure consistency, so whether that is positive or negative doesn't matter but rather that consistency does exist between the results
what is corrected ltem-Total Correlation indicate for?
Sounds weird though... Suppose your items (questions) were perfectly correlated, your alpha would be then >=1 (=excellent according to George and Mallery's interpretation). I would find this case undesiderable, it means you were not smart enough to "design" your questions; just knowing the answers of one question would let you perfectly guess the answers of the rest of the questions (?).
Imagine this case:
no! yes!
I love chocolate ☐ ☐ ☐ ☐ ☐
I hate chocolate ☐ ☐ ☐ ☐ ☐
Yes, you're right! In a good Likert scale, the individual items should not correlate extremely highly with one another!
Good Video..
Thanks for the nice feedback! Regards, Hannah
2:20 "the condition must be fulfilled... that all questions or items measure the same latent variable"
Does that mean that, I can't calculate cronbach's alpha on questionnaires which have a single item for every domain? Thanks!
mo😢
Thanks for making this video!! You made it so much easier to understand and love your examples!!!
Glad it was helpful! Many thanks for your feedback! Regards Hannah
This was actually VERY helpful - thanks so much for creating this!
Glad it was helpful! Regards Hannah
Would removing that 4th term to get a higher Cronbach's alpha result in loss or gain in power? What's to stop someone from removing another term or terms and remaining with only 2 or 1 terms?
You have the best explanations anywhere in the known universe of statistics! THANK YOU!
The best explanation I found after days of looking up various places for help. Thank You!
I will definitely have my students have a go at using your online statistical computing facility. Many thanks!
Many thanks Maria! Regards Hannah & Mathias
English from babble
Well explained!👏
Ah this is great. Danke aus Indonesien
I feel like I'm asking a stupid question but.... regarding the example you gave about extroversion:
because the "Cronbach's Alpha if deleted" for "enjoying social situations," is it too much to interpret that the item "enjoying social situations" is not a good indicator for measuring extroversion (in this specific example/dataset) altogether? Could it also mean that the other 3 items, though highly correlated to each other, are irrelevant to extroversion and "enjoying social situations" is actually the only appropriate item for measuring extroversion?
Sorry for the late answer! Yes, your sentence at the end is correct! With Cronbach's alpha, you can only measure the extent to which a group of questions is related to each other. Whether they really measure what they are supposed to measure is not something you can say! Therefore, it is also possible that all three items are very similar but measure something else and the one question only measures the "right thing". Regards, Hannah
@@datatab This was helpful, thanks!!
Thsi is sooo awasome
Your whole tutorials are amazing. A question: Do your DataTAB able to measure the validity (Discriminate and Convergent)?
what methods would you need for that? The simple correlation can be calculated of course! Regards Hannah
Madam, please add how to write in a dissertation about Cronbach's alpha statement
Many thanks for the feedback! I will put it on the to-do list! 😊Regards, Hannah
Thank you for making this look very less mystical. It was really helpful.
Glad it was helpful!
Thanks a lot. Excellent video!!! Easy to understand and useful.
Thank you so much, this video really saves my time.
Glad it helped!
Thank youu for the vid, very simple and comprehensive!
Glad it was helpful! Regards, Hannah
Thank you for the explanation and examples! It has been some time I didn't calculate Chronbach's a, so I forgot the way it works. That was clear and to the point.
Very precise and accurate...thanks alot 🙏
Thanks!!!!
great video. easy to understand. Thank you !!!
Glad it was helpful!
The alpha crombagh is used to find the realabiity of every kind of questions? thanks for you help
Need to calculate different variables captured in minutes how to apply for this
This is great thank you
You're very welcome!
hi, what is the minimum of cronbach's alpha? is 0.6 a good reliability?
Thanks so much. It is very clear explanation.
Excellent explanation. Thank you!
Many thanks! Glad it was helpful! Regards Hannah : )
Please can you also provide drawback of the test at the end it will really help
sweet voice
Many thanks! 😊Regards, Hannah
That was extremely helpful! Thank you!
Well explained!! Thanks
Glad it was helpful and many thanks for feedback! Regards Hannah
learned many new things in a simple way from here...the best veeos
Amazing explanation, thank you!
Many thanks! 😊Regards, Hannah
Thank you mam 😍😍❤️
Most welcome 😊
Thank you. Clear and sound
Many thanks!!! 😊 Regards, Hannah
MANY THANKS IT"s kind from you
Please can you carry out such test on questionnaires with yes or no answers?
In which situation , we can use the Alpha cronbach analysis
Nice, makes it easy understandble
Awsom thank you loved it
Well explained!❤
Thank you! 🙂
Great video!
Thanks!
An excellent explanation, thank you.
thanks so much
You're welcome and thanks for the nice feedback! Regards, Hannah
Thanks for the information.
Thank u!
You're welcome!
That is so useful!!!!!
This video is so great! It explains very clearly!
Great video. By the way, your voice remided me of audrey hepburn hehe:)
: )
Hello, I don't have an account of Data Tab, can you calculate the data for me?
It is cheap, just buy one!
@@datatab I don't have money, I am just a student, I'm from Philippines.
You sound like Arnold
: )
If I have 377 poeple to gather data, Respondents should be counted all of the people I have for Cronbach’s Alpha , is it right?
Thank you for explaining
Yes I think so, not sure if I understood the question exactly but that should fit!
@@datatab I have 29 questions and 377 respondents to interview, distribute questionnaire and 5- Likert scale,