How to analyze your Likert scale data in SPSS - Compute Procedure | Hotel Satisfaction Study

Подписаться 6 тыс.

Просмотров 20 тыс.

50% 1

#spss #kajidataonline #epiinfo #ppkp #ministryofhealth #ilkkm #hinditranslation #translation
Playlist: • Advance Diploma Emerge...
===FREE WORKSHOP Playlist Available below===
Join our TELEGRAM (t.me/kajidatao...) for LIVE Chat and Discussion about SPSS. You can post any question and enquiry, we usually conduct the LIVE discussion. Check out previous
Free 1 Hours Training 1: • How to run data analys...
Free 1 Hours Training 2: • Video
Syntax (Copy and Paste in Your SPSS)
* Encoding: UTF-8.
--- IV----
COMPUTE color=mean(a1,a2,a3,a4,a5,a6,a7,a8).
EXECUTE.
COMPUTE light=mean(b1,b2,b3,b4,b5,b6).
EXECUTE.
COMPUTE space=mean(c1,c2,c3).
EXECUTE.
COMPUTE decor=mean(d1,d2,d3).
EXECUTE.
--------DV --
COMPUTE satis.color=mean(aa1,aa2,aa3).
EXECUTE.
COMPUTE stais.light=mean(bb1,bb2,bb3,bb4).
EXECUTE.
COMPUTE satis.space=mean(cc1,cc2,cc3,cc4).
EXECUTE.
COMPUTE satis.decor=mean(dd1,dd2,dd3,dd4,dd5).
EXECUTE.
----- Normality Test ----
EXAMINE VARIABLES=color light space decor satis.color stais.light satis.space satis.decor
/PLOT BOXPLOT STEMLEAF HISTOGRAM NPPLOT
/COMPARE GROUPS
/STATISTICS DESCRIPTIVES
/CINTERVAL 95
/MISSING LISTWISE
/NOTOTAL.
==== Dataset ====
Click Here:
drive.google.c...
===============
=== IMPORTANTS LINKS===
*Subscribed Our Channel: bit.do/fmMvK
*Complete Playlist 1 (Workshop 2019): • Workshop SPSS Dec 2019
*Complete Playlist 2 (Workshop 2020): • RMC Workshop Oct 2019
*Free Online Learning Platform: spss.kajidataon...
*FREE ONLINE LEARNING: learn.myenviron...
*Buy domain and hosting less than USD50 per year, Get it Now: secure.gbnetwo...
*Support Us on Patreon (Buy me a coffee, Thanks bro): / kajidataonline
*The dataset for this exercise can be download freely here:
kajidataonline....
-----===----===---===---===---===---===---===---===
Do not forget to LIKE, SUBSCRIBE and FOLLOW us:
Facebook: / kajidataonline.spss
Instagram: / kajidataonline
Twitter: / kajidataonline
Website & Newsletter: kajidataonline.com
=====FREE WORKSHOP======
We also Offer FREE SPSS Workshop in Our Playlist, kindly visit to learn this 2-3 days workshop:
1) SPSS Workshop December 2019: • Workshop SPSS Dec 2019
2) SPSS Workshop October 2019: • RMC Workshop Oct 2019
3) Spatial Data Analysis Workshop: • Free Geospatial Analysis
4) Covid - 19 Data Analysis Workshop: • Covid 19 Data Analysis
Our Live Webinar Free (1 Hours Training): • How to run data analys...
If you need urgent help/ consultation for data analysis, contact us through Whatsapp:
spsshelper.wasa...
Admin Kajidataonline: kajidataonline.com

Опубликовано:

18 сен 2024

Ссылка:

Скачать:

Готовим ссылку...

Добавить в:

Мой плейлист

Посмотреть позже

Комментарии : 79

@arizafera7262 3 года назад

Thank you sangat2! Senang faham even for someone beginner. What you teach really saved my life! If not, I can't graduate this sem >.< Nanti kalau ada rezeki jumpa kat M'sia, confirm saya belanja makan!

@Kajidataonline 3 года назад

Alhamdulillah.. thanks for your comment puan. In sya allah, semua nya selamat hendaknya and good luck... share, like and subscribe. Thanks puan. Any help needed, please let us know: spsshelper.wasap.my

@funmiadesesan6111 3 года назад

This is great information. You described it in sufficient detail and supported the decisions made. I didn't know how to interpret the std. error of the skewness/kurtosis before watching this. Thank you so so much!

@Kajidataonline 3 года назад

Thanks for your comment. Please consider like, share it to your friend and do support us by subscribing and always follow us for latest update. Tq

@mihret-abmehari6765 4 года назад

This is so helpful. Thank you!!

@Kajidataonline 4 года назад

Thanks for your comment. Please consider to subscribe, like and share all our videos. Tq sir

@manjitkdhillon6476 4 года назад

this helped me aloT!....Muaks

@Kajidataonline 4 года назад

Thanks for your Comment. Please consider to subscribe, LIKE and share our content to your friend and family. Thank You for your support. For your Information, We offer a free webinar (With free E-Certificate). Check it out: ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-mpG9jekNwN4.html

@shahzad4ever1 4 года назад

Your video is very informative and educational and I love your videos, I have been in Malaysia, in Universiti Putra Malaysia for 3 years as postdoc fellow, Actually I am doing descriptive analysis on Linkert scale and little confused, I have 12 questions of knowledge, 13 questions about Attitude and 14 question about practices on street food, I want to analyze ANOVA and tucky test, could you help how to combine these questions

@Kajidataonline 4 года назад

Thanks for your comment. The tukey test is tukey kramers test for measuring the absolute differences for your anova test. The construct you have on your knowledge, attitude and practices can be analyze provided that your data meet the anova assumption. Please let us know, if you need more clarification. We would like to invite you in our telegram for discussion. t.me/kajidataonline_free and please do consider to subscrjbe our channel for more update.

@ringoinah 3 года назад

Dear Sr. first of all thank you for your many educational videos. I have a question: I am doing research on several farming types and farmers motivation and resilience. However some questions for example: what are your plans for your cattle with a score of 1 being no plans and a score of 5 being >4 plans. However if I was talking to a crop farmer this question did not apply and is registered as missing data. Can I still proceed with my calculations in SPSS and calculating minimum and maximum scores? Kind regards.

@Kajidataonline 3 года назад

Thanks for your question n comment. You can set the responses as missing values for sure. Just set the missing value indicator at the variable view of your variable. Usually researcher use the discrete number such as 999, or 9999. This indicator will instruct the spss to omit those responses and proceed your calculated variables as proper score. If your missing values was so obvious, I may suggest you recategorize/ recollect the data..

@joapriltaccad2899 2 года назад

Hello sir, I am looking for the next video here where in after you found out that the data is not normally distributed, it needs to run through non parametric test. Where is that video? did you use ordinal logistic regression? Can i have the link? Thank you.

@Kajidataonline 2 года назад

Thanks for your comment. I did the video about non-parametric analysis and logististic regression. For specific ordinal logistic regression, I may produce such video soon as you suggest.. anyway below is the link. Non-paramteric analysis: ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-8qj-iVlysFY.html Logistic regression: ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-s9j_qAer1nE.html If you got specific inquiry kindly watsup us for assistant (if available): spsshelper.wasap.my

@Usman_Khan786 4 года назад

Good Day! Your video is really very helpful in my academic purpose study and solve half of my problem.For the other half, your kind advice is required regarding data analysis. I am struggling with which type of regression analysis & statistical test I should use and HOW. How would I interpret the results to accept or reject the null hypothesis/alternate hypothesis. My question is I created a questionnaire in order to understand the relationship of DV (Social Media Ads features) and IV (purchase intention). I have one (1) dependent variable and one (1) independent variable Both DV & IV have Ordinal scale. Independent variables have 26 questions and dependent variable have 4 questions with a Likert scale of 5 for each question. (strongly disagree=1, disagree=2, neutral=3, agree=4, strongly agree= 5) Can you please suggest me the right procedure. Which type of regression analysis or statistical test that I suppose to perform and how, that may satisfy the requirement of my current study. Thanks in advance,

@Kajidataonline 4 года назад

Dear sir, thanks for your comment. Perhaps, you can share your idea/ question with us. Just email, and we are happy to review and comment. Email : admin(at)kajidataonline.com You can also join our telegram for rapid communication. t.me/kajidataonline_free

@ClaraJaborandy 4 года назад

Thank you very much!

@Kajidataonline 4 года назад

You're welcome! Please consider to subscribe our channel. Thanks

@nurhusnani98 3 года назад

This video is really helpful. Can I know if the data is not normal since it is on likert scale, can we proceed to do the factor analysis? Or need to transform first? But I already did the transformation, but the data still not normally distributed.

@Kajidataonline 3 года назад

Thanks for your question. The data can be transform two times, and if the data still not showing any improvement of normality, then you can proceed with non-parametric analysis

@nurhusnani98 3 года назад

@@Kajidataonline okay thank you sir. Have a great day!

@salsabila-xo4bq 4 года назад

is it possible if we use ordinal regression for this type of data? dependent and independen variables are ordinal data too.. thx

@Kajidataonline 4 года назад

Yup.. thanks for your comment. Surely, ordinal regression can be deploy for this example. However, the use of construct by mean should be very easy to interprete and discuss. Anyway, its depends on the objective. If ordinal regression want to be use, ensure your dv was very important to be address

@souravmukherjee01 3 года назад

Do you have any video (with data) showcasing 6-7 IV and just 1 DV? Also with some demographical data but those are not contributing to it....just taken for analysis?

@Kajidataonline 3 года назад

Thanks for your comment. Please have look our past webinar for the complete data analysis. It might be lengthy, but I believe the example and demonstration are complete. Please subscribe to our channel as we will go to explore the analysis step in SPSS. Link: ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-3vNz2Gm7JRU.html

@e30e82 3 года назад

Hello, Thank you for the explanation. I have a question. My dependent variable consists of three items, and I decided to compute them into one variable. I used a 5-point Likert scale. The output for the mean is, for example, 4.3. How is it related to the 5-point Likert scale? Is it possible to have the same range? Or how shall I interpret the mean of 4.3, based on the 5-point Likert scale?

@Kajidataonline 3 года назад

Yes.. your 5 likert scale usually interpret based on your 5 stages.. therefore the mean of 4.3 usually suggest as agree side of the likert questionnaire. However, just a simple taught, please do look into your standard deviation for their dispersion. Worth looking for minimum and maximum to understand the pattern of your data..

@sofiap318 3 года назад

Hi! Can I ask ... I have 3 Likert items measured on a 5-point scale and I computed the mean of these items to get a variable. This variable is my dependent variable and I'd like to use it in a regression where my independent variable is of the same type, so is the result of the mean of 3 likert items. What type of regression approach should I use? As my DV is the mean of multiple items, I think I can assume the data are interval data and so go for linear regression. In fact how could I do an ordinal regression with this type of DV? Thank you

@Kajidataonline 3 года назад

If you summated mean already your dv. You can proceed with linear regression or multiple linear regression..

@burcugumus3047 3 года назад

Thank you for the video. It is really helpful. But I have a question :) Let’s think the questionnaire contains 4 categories and each category has 5 questions. When we transform the categories based on the mean value , as in the video, we create a new data set indeed. So is there any other way to combine different questions into one category?

@Kajidataonline 3 года назад

Thanks for your comment. The way to compute the mean of the score from categories and question are generally based on literature and construct of the questionnaire. In theory, the way the calculation for categories are coming from the exploratory factor analysis. Your question, the way to combine as one category basically compute using overall questions as one holistic score, but of course not meaningful in term of understanding the pattern.

@abiramisinnan4604 3 года назад

Hi sir, if swekness value is 0.56 and kurtosis is 2.01 is it needed to run ANOVA or kurskal walis

@Kajidataonline 3 года назад

Thanks for your comment. If that is the value of statistical skewness, take your Standard Error (SE) skweness then times by 2, if your value of statistical skewness is LOWER than your SE x 2, that means the data was normally tabulated, and therefore you can proceed with parametric analysis (ttest anova etc), if the value higher than that, its not normal

@835tej 4 года назад

I have my questions in Likert Scale: Strongly Agree, Agree, Neither agree nor disagree, Disagree & Strongly Disagree. I tried using the Mann-Whitney test on SPSS but it doesn't consider all the 5 options. How do I analyze the data? Your help would save my life...

@Kajidataonline 4 года назад

ok, firstly, in research usually we graded with the negative towards positive. The next comment is, the analysis you trying to do must consider any differences test for what basis. I meant, the questionnaire you have probably contain 5 question that represent one construct, let say, 5 question measure stress, this stress level will be compute using summated mean or cumulative score algorithm (depends on what literature you based on). From that score you proceed to measure differences between variable e.g. gender and so on. If you look into my video, the data firstly need to be identify the construct, calculate the mean or sum of the construct then proceed with the analysis. If you trying to just see individual question, you can do it also, but surely its not fully correct.

@lindaayurizkaputri6535 3 года назад

This is useful video indeed. great thanks for posting this video. somehow, I have a question, I am going to analyse my data on the effect of health education to the motivation. The motivation consist of 7 questions measured in likert scale ( I assume this is ordinal scale), and I made 2 measurement (pre-post measurement). how do you think I should analyse my data. thanks in advance.

@Kajidataonline 3 года назад

Thanks for your comment. You can use the paired sample t-test. If you conduct any experiment where measuring the differences of time and group (as example), you can use factorial anova.

@lindaayurizkaputri6535 3 года назад

Thank you for considering my questions, that's very kind of you. So I assume that the motivation is treated as numerical variable. I was confused about the data scale, numerical or ordinal given that the questions were in likert scale

@Kajidataonline 3 года назад

When you use algorithm to calculate mean or sum of the score, the ordinal now transform into interval scale. Its not ratio.. therefore in spss, any scale (interval/ ratio), can be consider numerical analysis for mean differences and so on

@ainfatini7076 3 года назад

hii can I ask, i have 5 option in likert scale. and I want to measure chi square every section of my question with socio demographic. how i want to key in likert scale question in spss after i make pooled response every section for each liker scale response.

@Kajidataonline 3 года назад

Thanks for your comment. Clearly, i did not really understand what you are trying to do.. can you whatsapp me instead? And perhaps can share the data, questionnaire, if its simple one, can produce free tutorial. Email: admin(at) kajidataonline.com AND whatsapp: spsshelper.wasap.my

@aaronayubasamaila8821 4 года назад

Nice video. Can I ask? I have 2 independent variables and 1 dependent variable. Each time I run the syntax generated it only shows me that of the 2 independent variables while the dependent isn't displayed in the data view. What could possibly be the problem? Anticipating your honest reply the soonest. Thanks.

@Kajidataonline 4 года назад

Thanks for your comment. I may not clearly possible to solve without having the picture of your syntax. You can share in our telegram group. t.me/kajidataonline_free I would like to suggest a solution possibly due to your split file/ filteration command you had been run previously. My suggestion, try look into your syntax and ensure split file off is activated. Thanks

@aaronayubasamaila8821 4 года назад

@@Kajidataonline Ok sir. I will screenshot and then get back to you.

@Kajidataonline 4 года назад

Thanks

@aaronayubasamaila8821 4 года назад

Hello sir! I noticed a problem. I imported from excel and instead of getting data for 286 respondents, SPSS keeps the respondents at 285. What could possibly be the problem?

@Kajidataonline 4 года назад

@@aaronayubasamaila8821 Dear Sir, have you consider check your title of variable or "Upper Part of Excel is the Variable Name", I believe, the lack of one data will be the first row in excel usually contain the variable name. If you want to copy from excel to SPSS, please make sure you select the "data" only not including the name of the variable.

@emeliazulkifle9287 4 года назад

hi! can i use the same computing score method and proceed with multiple regression?

@Kajidataonline 4 года назад

Yes sir. You can use the regression method straight away from syntax interfaces.

@khells2375 4 года назад

Hii can i ask, i have 6 option in likert scale, how do i actually merge the option together without having to do mean median etc i just want to combine 5 ques and see all of those 5 ques answer

@Kajidataonline 4 года назад

Thanks for your comment. I am happy to help, maybe you can share the basis of your data, I can demonstrate to you. The idea was using recode and also analyze multiple responses. Tommorow we will have the live session for data analysis, perhaps you can join

@fastuser007 4 года назад

how can you find mean of ordinal data? you should calculate median and mode. you are wrong.

@Kajidataonline 4 года назад

Thanks for your comment. For your information, the raw nominal scale yes, you cannot run the mean, but summated mean/ computational method wise, you can, because the SUM of overall construct will be interval scale, and eligible for running the mean and SD

@veronika2315 4 года назад

Hi, are the age group and income group always nominal or ordinal in SPSS?

@Kajidataonline 4 года назад

By default, its depend how you set. If you set your questionnaire with the option of range e.g. range 30, this data will generate nominal scale of measurement, similarly with income. The ordinal as definition will be similar as nominal except, its got ranking in term of the staging. Anyway, scale of measurement should be divided into two major categories (alpha numeric and numeric), alpha numeric will be nominal and ordinal (the numeric values given to the system in order to "teach" the computer understanding of the attribute). Another numeric will be interval and ratio

@AsfikaSilvi 4 года назад

Hello Sir, I am doing a likert 5-scale survey research on the topic of supply chain management practices and performence of a company where practice is the independent variable and performance is the dependent variable. In the questionnaire there are 15 questions related to practice and 7 questions related to performance. Can you please suggest me what type of analysis I can do and how in my research using spss?

@Kajidataonline 4 года назад

Thanks for your commnet. Based on your question. I would like to suggest you to use the summated mean or total score by each construct. The construct will be define as collection of your question for IV and DV. However, the construct firstly need to be explore and establish during exploratory factor analysis by looking into similarities. Those 5-scale can be compute using similar approach as we suggest in our video

@AsfikaSilvi 4 года назад

@@Kajidataonline Thanks for replying. Can I do multiple regression and one sample t test for the analysis? if so, then how? I am really confused. I would be really grateful if you help.

@Kajidataonline 4 года назад

I can help, surely, if the data already organize. Maybe you can share to us for discussion. Please whatsapp for more details. spsshelper.wasap.my Please also consider joining us in our telegram: t.me/kajidataonline_free And you can email the data and objectives. Email: admin[at]kajidataonline.com

@AsfikaSilvi 4 года назад

@@Kajidataonline Sir I have sent you the spss data and questionnaire through mail. Thanks.

@Kajidataonline 4 года назад

Will have a look and discuss later

@munirahnordin5644 3 года назад

Tuan, nak tnya, apa test yg sesuai untk cari significant factors? Objektif study sy adalah : to identify the significant factors that affecting the management of emerging risks in SMEs ... sy ad 5 factors yg dikaji... setiap jenis management tu sy mintak responden rate kan setiap 5 faktor tu menggunakan likert scale(strongly disagree-strongly agree)... harap tuan dapat bantu sy☺️

@Kajidataonline 3 года назад

The best idea for un structured construct will be factorial analysis.. after run the efa, can proceed with validation using confirmatory factor analysis (cfa)

@munirahnordin5644 3 года назад

Kalau taburan data x normal, still boleh guna descriptive analysis x?

@Kajidataonline 3 года назад

Thanks for your comment. Ya puan.. data descriptive masih boleh guna.. even data normal atau tidak.. beza adalah jenis analysis/persembahan data yg dikehendaki. Cth. Data normal utk data numeric, kita akan pakai mean dan sisihan piawai.. tetapi jika data tidak normal, kita akan gunakan median dan interquartile range.

@munirahnordin5644 3 года назад

@@Kajidataonline baik.... tq tuan☺️

@Kajidataonline 3 года назад

Thanks puan

@lanadellover9349 3 года назад

Hello, I'm struggling to decide exactly which statistical analysis that I need to run on data collected for my thesis. Can I send you my hypotheses so that you can guide me accordingly? What's your contact info?

@Kajidataonline 3 года назад

Thanks for your comment. Please consider to whatsapp us at following link: spsshelper.wasap.my

@lanadellover9349 3 года назад

@@Kajidataonline I have whatsapped you over 24 hours ago yet no reply.

@Kajidataonline 3 года назад

Thanks.. We are sorry, we are out from city areas, therefore the internet connection was not good. We currently conduct environmental studies in the forest area. Will be available in few hours.

@7918476 3 года назад

@@lanadellover9349 calm your entitled a$$ down. He doing you a favor.... I would charge you for being so impatient

@Kajidataonline 3 года назад

Dear Lana, the statistical analysis, basically for hypothesis (divided into 2/3 types, the hypothesis for differences, and hypothesis for correlation/ relationship and the third will be the modelling. So how to decide, basically depends on the objective/ what you want to do.