Dear Shaker,

If you adopted questionnaires from previous studies (that used 5-likert scale) a good reason to keep the scale is the comparability of the results. Usually when we replicate a research we wish to compare results.

its a proven scale for most of the researches in Social studies

thank you brother but the thing I must provide based on previous researchers and explain the important of using such scale. not only because a common use then I need to follow.

Can i upload a article related to the usage of likert scale

Yes please

Because the middle answer, if it is 'no opinion' or 'don't know' or 'neither a nor b', can tease out whether the question is actually asking for the correct information in a well targeted manner. Lots of responses in the middle category would indicate a poorly phrased question, or a question that has multiple possible answers that means respondents are unable to choose, and few answers in that category might indicate a well targeted question.

just paste this link

http://jalt.org/test/PDF/Brown34.pdf

Whether the link is useful?

thank for both of you brothers, I still cant get the idea, I need a clear sentence like the following sentence which I found in one thesis but unfortunately it was about 7- likret scale :

Respondents were asked to respond to the items by indicating their level of agreement using a seven-point Likert scale. This category of scale was chosen because based on Hair et al. (2007), respondents usually avoid the extremes in responding to the questionnaire items. So if there are seven scale responses, the instrument actually has

only five-point scale. This is due to the fad that it is highly unlikely forrespondents to answer the most extreme scale (i.e. 1 or 7).

I know Suzan, but the problem my examiner asked me to explain why This category of scale was chosen in my study.

You can use a 5 point likert scale or any other scale you like ( for example a 7 or a 9 point likert scale). However, your choice may be due to 2 reasons. Number 1: it's much easier for you to analyse the data and for the people you will interview to stay focused on your questionnaire and answer you with more honesty and criteria. However, reason number 2 may be due to what is wrtitten in the literature. If any authors have applied the same question in similiar ( or also different) studies you should mention it and you should also try and use that methodology in order to compare results and add your views. I hope I contributed towards your needs and I wish you the best for your work.

Dear Shaker,

If you adopted questionnaires from previous studies (that used 5-likert scale) a good reason to keep the scale is the comparability of the results. Usually when we replicate a research we wish to compare results.

The Likert scale from 1 to 5 have a midle-point. Respondents also find it difficult to define their views on scales with values ​​greater than seven. This means that if you offer more than 7 answer choices, respondents are likely to choose their answers at random and thus, their data will be insignificant. It is recommended using five-point scales for single scales (the presence or absence of a selected attribute is referred) and seven-point bipolar scales (Includes two sensations or different attributes related to each other).

Thank you much guys. i just need to share with u all what i have give justification regarding to this issue and if there is an improvement please share with me.

i have justify as the following:

Respondents were asked to respond to the items by indicating their level of agreement using a five-point Likert scale. five-point Likert scale used because this kind of scale can make a compromise between the conflicting goals of offering enough choice (since only two or three options means measuring only direction rather than also strength of opinion) and making things manageable for respondents (since few people will have a clear idea of the difference between, say, the eighth and ninth point on an eleven-point agree-disagree scale). In addition, it is mostly recommended from previous studies (i.e. Fink, 1995; Dawes 2008; Dillman, et al. 2009).

Mr Shaker, I completely agree with Madalena Ramos thought. I think you should justify saying that prior studies in your research area used 5-likert scale since you have adapted from past studies. It is a norm to use the instruments that are inline with current research trend. Therefore,to be consistent and compare your findings with prior studies, you have right to use prior studies instrument and scaling .

:) Yesss thank you brother

Because that seems to be the standard or norm for social science research. There are also studies which suggest that the larger the scoring scale used, the greater the variability.

Shaker, I think Madalena Ramos gave you the best answer so far. That is, if you're adapting from previous work comparability and consistency are valid justifications. In my own experience I've always preferred odd numbered scales, typically with five choices for a couple reasons. It's also fairly close to a "standard," and makes answering easier on your respondents.

First, the odd number includes a clear mid-point that I've always referred to as the indifference point, that is, without strong preference in either direction. Second, too many points do not necessarily add any value or increase the validity of your findings. There's a good reason behind this too, also based on my experience.

Assuming you have a large enough number of respondents your results are going to converge on a mean score in any case. And the mean, along with the distance of responses from the mean are the bases for lots of analytical approaches. In my experience it doesn't really matter how your scale is numbered, eventually the mean is going to be pretty much the same. I had a years-long experience that convinced me of this tendency.

A state agency sponsored research into its clients' satisfaction with its programs (technology-driven economic development) that I led for seven years. I initially used a five-point scale. The client argued it did not give respondents enough "fine" choice. So I went to a seven-point scale. Same argument from the client, who eventually insisted on a 100 point scale. Not only did the 100 point scale look ridiculous on the questionnaire, but there was no significant difference among the mean scores. I typically had several hundred responses.

So, a large enough N is going to result in roughly the same mean, and the mean is the sort of "magic" number you need to analyze your findings. Thus, that fact combined with Ramos' advice is your justification, I think.

Nice answer dear James thank you much

Dr. Jacob L. Oyugi

I join the debate by supporting some of the reasons given by James especially the issue of mid-point. It is like a number line where 0 separates negative and positive integers. A five-point scale shows strength of agreement on one side of the scale and strength of disagreement on the other side. Some respondents are indifferent in some cases with their degree of agreement and therefore have no choice except to take the mid-point. The fact that it was used in another study is not a strong justification because the question still remains unanswered.

In addition to the reasons given in some of the previous answers, I usually also mention the Miller's law, as described in http://en.wikipedia.org/wiki/The_Magical_Number_Seven,_Plus_or_Minus_Two

Which states that the number of objects an average human can hold in working memory is 7 ± 2. This justifies a 5 to 9 point scale. Within the range, more points give more sensitivity in measurement but perhaps make it more difficult (and take longer) for the respondent to think about the answer. It is up to the researcher then to justify the best trade-off.

May you can find more details in the attachment.

There are a variety of scales. One can use Likert Scale of three, five, or seven point. In the literature, there are reasonable reasons and supported justifications on using the one with five points.

Thx much Gary Bourque

Hi Shaker,

If you want, you can refer to my answers (2 posts) toward this similar RG question:

https://www.researchgate.net/post/When_should_one_switch_to_the_10-item_Likert_scale

Basically I explained what are the advantages & disadvantages of larger scale vs smaller e.g. 10pt- vs 5pt-Likert Scale as well as how to convert the Likert Scale to Semantic Differential scale using Bipolar Adjectives (e.g. Effective1, 2, 3, 4, 5, 6, 7 Not Effective) in order to  use parametric statistics.  All the best.

Regards,

Fung

The most recommend scale is the one of 5 points, where it is reliable than 3 or 7 points. There are many justifications to support this. You can consult previous studies on this. 

The five-point Likert scale type of questionnaire will force the respondents to make a choice, since the neutral option is not available, as long as the researcher will have appropriate choices using the ordered continuum response categories.