Nashbramhall
Learning & Teaching Fellow (retired)
Mnsvasan
Accounting, Finance
Tomar_anjita12345
Assistant Professor
+3 Others

Thread Started by #sajjal20

Hi:
I recently conducted a survey in which I utilised the 5 point likert scale to gather data. The only problem is, I dont know how to analyse it. I feel like a complete idiot for not having considered this before. I thought it would be easy, except I dont know how to write an analysis. Can someone please help? It would be really useful if a sample was provided with an analysis of a 5 point likert scale.....
Thanks in advance!
LATER ADDITION:
I have, belatedly, attached a sample of my writing to the post. People, please comment on the appropriateness of the analysis. The attachment is in one of my posts below! Thanks!
17th April 2011 From Pakistan, Rawalpindi
Please search the web using keywords "5-point Likert Scale data analysis". You will find links which give instructions as to how to analyse such data.
Have a nice day.
Simhan
A retired academic in the UK
17th April 2011 From United Kingdom
Likert scale points – it is an attempt to quantify the subjective views.

For each point there should be a meaning for each number. Eg. for 1 = most objectionable. 2 = objectionable…. 5 = most desirable.

You must use these expressions in your analysis or presentation of the interferences. We may state out of 1500 persons studied (sample) 200 strongly objected to the proposal and 20 considered the proposal most desirable.

We must remember that we are not handling any data about a problem. We are handling the number of persons who have taken a particular view. Any mathematical process of mixing this data with a data on a particular problem should be designed carefully.

If you have included many questions using likert scale, you may tabulate the views carefully seeing if these are logically connected. Your presentation can have a statement like “72% of those who “Strongly Opposed” to issue A, Mildly Objected issue B also. Or
55% those who welcomed production based incentive strongly objected to change in shift timings.

There are several published charts and reports in the net – use them as a model or starting point – it is easy. I have often visited “talkstats.com” and found the site helpful.

You may also try.

Good luck.

Srinivasan
17th April 2011 From India, Madras
Hi:
I have searched that...but most of the results that popped up went into deep statistics, and that freaked me out a bit.
I was hoping that you could probably read the analysis of one of my charts, and give your valuable feedback on it, since I have been unable to acquire a sample report relevant to my topic.
18th April 2011 From Pakistan, Rawalpindi
Here is the attachment! I am new here, so its a little confusing how to manage the posts....will hopefully cotton on soon enough! :)
18th April 2011 From Pakistan, Rawalpindi

Attached Files
Membership is required for download. Create An Account First
File Type: pdf sample.pdf (30.6 KB, 443 views)

Hi!
As Srinivasan has rightly said, Likert scale points is an attempt to quantify the subjective views. Obviously, its a Part of Statistical studies.
Your pdf sample is good show for the first attempt.lease do not get overwhelmed by the tests being used at this stage. But, otherwise, there is no shortcut for success.
All the best.
18th April 2011 From India, Mumbai
Dear Sajjal,

Thank you for attching an example. However, how you analyse the data depends upon your objectives, the amount and nature of data that you have collected, and the level of the course. For example, if you are an undergraduate, a simple pie chart or bar chart and percentage distribution would do. If you are studying for a post graduate degree, you will be expected to exhibit your knowledge of the application of statistical tests. Similarly, if you have collected demographic data, you may be able to see if there is any difference between males and females or responses from different work areas, levels, etc.

Also see <link outdated-removed> to see how a student has given details about the research's conceptual framework, etc. Though (s)he has not used a Likert Scale, (s)he has used both a questionnaire survey and interviews to TRIANGULATE the findings. The project report is not one of the best I have come across but much better in the quality of literature reviews, methodology, and the list of references, etc.

Have a nice day.

Simhan
18th April 2011 From United Kingdom
Hi Simhan:
Thanks for your reply! I am working towards an undergraduate degree. I am guessing the piechart should then suffice! The sample size is 40, subject is employee motivation, and the degree I am working for is not focused on HR.
Kindly provide some feedback with regard to the content written....it would be of big help! Any recommendations would be highly welcome!
Sajjal
18th April 2011 From Pakistan, Rawalpindi
Hello:
Yes, I agree, likert is a part of statistical studies, but since my degree has not drawn in any hard core statistical calculations, I am assuming piecharts and bar graphs is the level expected.
Thank you for your feedback on the sample....I wonder if you would be kind enough to point out the areas it was deficient in, so I can make improvements accordingly.
Thanks!
Sajjal
18th April 2011 From Pakistan, Rawalpindi
If your intention is just passing, what you have done by way of analysis will suffice. Whether you will get a grade better than a pass will depend upon how you will write the other chapters (Introduction, literature review, methodology, conclusion and recommendations).
18th April 2011 From United Kingdom
Hi Sajjal,
Your effort if good and will give you the required result. Group the questions on areas like communication, leadership, motivation etc and check the averages. You can get into the depth og these groups (question wise analysis) later on and highlight those which are of concern. Also make sure you give recommendations in your presentation.
Hope this helps
18th April 2011 From India, Visakhapatnam
Hi! Thanks for your reply. Thats a great idea! Didnt consider grouping the like questions together and giving an overall picture. Thanks! It definitely helped :)
19th April 2011 From Pakistan, Rawalpindi
Grouping like questions is a good idea; however it should be based on statistical tests. As you are an undergraduate, and may not have been taught, you can mention the limitations of your questionnaire. Please see How to Create Survey Questions | eHow.com and Qualities of a Good Survey Question
Have a nice day
Simhan
19th April 2011 From United Kingdom
Hi Simham:
Thanks for the links, they were quite informative.
I have actually formulated the questionnaires based on a couple of motivational theories. I was thinking of merging the like questions under the common concepts by each theory, and analyse them accordingly. That way I could back up the analysis with theory concepts. Would that be ok?
Thanks for your help.
Sajjal
19th April 2011 From Pakistan, Rawalpindi
It would be ok. Please drop me a private message. I will give my emailID and you can send what you have written so far and I will givew you feedback.
19th April 2011 From United Kingdom
In my opinion, writing survey report for the results shown in sample is rather easy. But if you are interested in writing a solid poicy oriented report, you have to bifurcate these results in various sub-categories (worker, supervisor, manager - age groups 25-30, 35-40, and gender wise male/female). It will give you handle and strong conclusion that if some people disagree with a proposal, who they are and what type of policies should be adopted to satisfy them or how include them in the process of change.
Good Luck!
20th April 2011 From Pakistan
Hi: Thanks for your reply! You know, thats a really great idea! I will have to type my stats in again on excel, but I think it will be worth it! Thank you!
20th April 2011 From Pakistan, Rawalpindi
Moinkhan's idea is good provided you have gathered info from enough people. As your total responses is only 40, commenting on differences among different age groups, postions, etc., may not be statistically valid. You need to apply test to see if the differences are significant. Hence, make sure that you put a note to that efdfect in your analysis or conclusion sections.
Have a nice day.
Simhan
20th April 2011 From United Kingdom
Simhan, could I please get your email address? I can email all the details to you so I dont have to post it in bits.
20th April 2011 From Pakistan, Rawalpindi
Hi, my data includes students' grades (X, XII and college) and their responses for different factors like study habits(3 questions), personality traits (5 questions) etc on 5-point scale (5 – Always, 4 – Often, 3 – Sometimes, 2 – Rarely, 1 – Never)
My objective is to study "factors affecting academic performance of students"
1. Should I consider median for each question and then compare it for male and female group by using chi-square test?
2. How can I measure the effect of these factors on students grads?
Please help. I am stuck ..
16th February 2016 From India, Gurgaon
Have you tabulated the results in Excel table. If so, could you kindly send that to me (Please see my private message). I will take a look and advise you.
Kindly also read the contents at What statistical analysis should I use? Statistical analyses using Stata and https://www.quora.com/When-should-on...sed-to-average
More links at https://www.google.co.uk/webhp?sourc...an%20or%20mean
16th February 2016 From United Kingdom
Reply (Add What You Know) Start New Discussion

Cite.Co - is a repository of information created by your industry peers and experienced seniors. Register Here and help by adding your inputs to this topic/query page.
Prime Sponsor: TALENTEDGE - Certification Courses for career growth from top institutes like IIM / XLRI direct to device (online digital learning)





About Us Advertise Contact Us
Privacy Policy Disclaimer Terms Of Service



All rights reserved @ 2019 Cite.Co™