Interpreting Statistics Report for Ranking Question

Plus d'informations
il y a 1 an 7 mois #161969 par blocka
I created a ranking type question with 6 answers (question code Q2).

When I go to the statistics section and include that ranking question to review, I get 4 tables -- like Q2[1], Q2[2], Q2[3], Q2[4].

I don't understand how to interpret these tables (Ranking 1, Ranking 2, Ranking 3, Ranking 4). The answers are listed in the same order in each, but with different counts and %'s.

What does this mean? How I can tell what was ranked highest overall? Or what the overall ranking of each answer was?
Les utilisateur(s) suivant ont remercié: ventolinmono

Connexion ou Créer un compte pour participer à la conversation.

LimeSurvey Partners
Plus d'informations
il y a 1 an 7 mois #161973 par holch
Can you show a screenshot of what you are seeing?

I'm not a LimeSurvey GmbH member. I answer at the LimeSurvey forum in my spare time. No support via private message.
Some helpful links: Manual (EN) | Question Types | Workarounds

Connexion ou Créer un compte pour participer à la conversation.

Plus d'informations
il y a 1 mois 1 semaine #185501 par ventolinmono
Hi

I have the same question. I get several and different results from the same ranking question on expert mode.
Attached a screenshot.

Thank you.
Pièces jointes :

Connexion ou Créer un compte pour participer à la conversation.

Plus d'informations
il y a 1 mois 1 semaine #185502 par tpartner
The ranking type question creates a data column for every possible rank. So it essentially becomes a series of questions. In the data you will see all of the answers ranked in position 1 in one column , all of the answers ranked second in another column, and so on. This leads to the different charts for each possible rank position.

Cheers,
Tony Partner

Solutions, code and workarounds presented in these forums are given without any warranty, implied or otherwise.
Official LimeSurvey Partner - partnersurveys.com

Connexion ou Créer un compte pour participer à la conversation.

Plus d'informations
il y a 1 mois 1 semaine #185503 par ventolinmono
OK. And which answer is the most common or frequent? Would that be the first table?

Connexion ou Créer un compte pour participer à la conversation.

Plus d'informations
il y a 1 mois 1 semaine #185508 par Joffm
First: The LimeSurvey statistics is just to have a quick glance at the results. Here you see which objects were ranked first with which frequency, or ranked last. All further calculations you will do in your statistic tool.

And which answer is the most common or frequent? Would that be the first table?

There is no simple answer.
Let me explain by ranking products.
You get this data table.

The first column shows the products ranked first.
You see "Product 3" is ranked first 3 times, more than all others.
But is it the best product? It is ranked last 3 times, too. So the opinion about this product is very ambivalent.

To analyze your results, this way to display the ranks is not the best.
To find out the overall rank of each product you have to calculate the middle rank, the mean of the ranks.
So in your analyzing tool you should restructure the table to

The columns now show the rank of one product.
Now you can calculate all statistical tests to find out which one is ranked best, is there a significant difference between the products
As shown you might get the same mean for several products
There are products which are either loved or hated (many times first place, but also many times last place)
Other products are more or less boring, uninteresting (ranked in the middle)
Now you may calculate the standard deviation of the product ranks.
Here you see that the opinions about "Product3" are more equivocal than about "Product2"

And even if you see a difference ("Product1" vs. "Product2") this might be of no significant meaning.
The t-test shows this.


Joffm


Volunteers are not paid.
Not because they are worthless, but because they are priceless
Pièces jointes :
Les utilisateur(s) suivant ont remercié: ventolinmono

Connexion ou Créer un compte pour participer à la conversation.

Plus d'informations
il y a 1 mois 2 jours #185868 par ventolinmono
Great answer, thanks. I'll try this.

Connexion ou Créer un compte pour participer à la conversation.

Commencez dès maintenant !

Créez simplement un compte et commencez à utiliser LimeSurvey dès aujourd'hui.

Inscrivez-vous maintenant

Inscrivez-vous à notre Newsletter!