- Posts: 15
- Thank you received: 0

# DECIDE THE GROUPS ORDER

3 years 1 month ago #112649
by GiuseppeC

GiuseppeC created the topic: DECIDE THE GROUPS ORDER

Hi, I have a question and I couldn't find any solution.

If in my survey i have two GROUPS OF QUESTION

1) GROUP OF QUESTIONS 1

2) GROUP OF QUESTIONS 2

I want that 10 respondents answer first the GROUP OF QUESTIONS 2 and than the GROUP OF QUESTIONS 1 and 10 respondents viceversa (WITHOUT RANDOMIZATION)

has anyone a solution?

Many thanks

Giuseppe

If in my survey i have two GROUPS OF QUESTION

**in this order**:1) GROUP OF QUESTIONS 1

2) GROUP OF QUESTIONS 2

I want that 10 respondents answer first the GROUP OF QUESTIONS 2 and than the GROUP OF QUESTIONS 1 and 10 respondents viceversa (WITHOUT RANDOMIZATION)

has anyone a solution?

Many thanks

Giuseppe

Please Log in to join the conversation.

3 years 3 weeks ago - 3 years 3 weeks ago #112816
by GiuseppeC

GiuseppeC replied the topic: DECIDE THE GROUPS ORDER

I explain better:

I ask the gander in a filter question (male-female)

I want, for example, that male respondents answer first GROUP OF QUESTIONS 1 and then GROUP OF QUESTIONS 2

while female respondents answer first GROUP OF QUESTIONS 2 and then GROUP OF QUESTIONS 1

Has anyone a solution please?

I ask the gander in a filter question (male-female)

I want, for example, that male respondents answer first GROUP OF QUESTIONS 1 and then GROUP OF QUESTIONS 2

while female respondents answer first GROUP OF QUESTIONS 2 and then GROUP OF QUESTIONS 1

Has anyone a solution please?

Last Edit: 3 years 3 weeks ago by GiuseppeC.

Please Log in to join the conversation.

3 years 3 weeks ago #112820
by holch

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

holch replied the topic: DECIDE THE GROUPS ORDER

I can only think of creating group 1 twice.

Then you can use conditions to hide the ones that should not be shown based on the gender of the respondent.

EG: Male should see first group 1 then group 2, females should first see group 2 than group 1.

So you make a copy of group 1 and put it after group 2, call it group 1f for "female" or whatever you want.

Then if respondent is female hide group 1, so they will fill in first group 2 and then the copy of group 1 after group 2.

When respondent is male, hide group 1f.

Then you can use conditions to hide the ones that should not be shown based on the gender of the respondent.

EG: Male should see first group 1 then group 2, females should first see group 2 than group 1.

So you make a copy of group 1 and put it after group 2, call it group 1f for "female" or whatever you want.

Then if respondent is female hide group 1, so they will fill in first group 2 and then the copy of group 1 after group 2.

When respondent is male, hide group 1f.

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

The following user(s) said Thank You: GiuseppeC

Please Log in to join the conversation.

3 years 3 weeks ago - 3 years 3 weeks ago #112823
by GiuseppeC

GiuseppeC replied the topic: DECIDE THE GROUPS ORDER

Thank you, I've already thought about this.

The only problem is that the result file will have 4 different groups of questions, 2 for male and 2 for female. So I have to reorder it in Excel later.

Is it possible to change the destination of the answers in the result file (maybe with JAVA) in order to have only 2 groups of questions in the result file?

I hope I explained it clear

The only problem is that the result file will have 4 different groups of questions, 2 for male and 2 for female. So I have to reorder it in Excel later.

Is it possible to change the destination of the answers in the result file (maybe with JAVA) in order to have only 2 groups of questions in the result file?

I hope I explained it clear

Last Edit: 3 years 3 weeks ago by GiuseppeC.

Please Log in to join the conversation.

3 years 3 weeks ago #112829
by holch

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

holch replied the topic: DECIDE THE GROUPS ORDER

I understand what you are trying to do.

First of all: with my solution there would not be 4 different groups but only 3. You can reuse the second group.

And I guess there might be some possibilities to do something with Javascript (not JAVA, totally different) or with Expression Manager.

But fixing this in Excel should take you less than a minute (filter by gender, copy the columns to the other empty columns- ready) that it isn't even really worth to think about another solution, if you do not need the answers for some filtering or calculations later on in the survey.

You could for example write the results from the response that might have two groups into a equation question. But I think you will still need the solution with group 1 before and after group 2.

First of all: with my solution there would not be 4 different groups but only 3. You can reuse the second group.

And I guess there might be some possibilities to do something with Javascript (not JAVA, totally different) or with Expression Manager.

But fixing this in Excel should take you less than a minute (filter by gender, copy the columns to the other empty columns- ready) that it isn't even really worth to think about another solution, if you do not need the answers for some filtering or calculations later on in the survey.

You could for example write the results from the response that might have two groups into a equation question. But I think you will still need the solution with group 1 before and after group 2.

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

Please Log in to join the conversation.

3 years 3 weeks ago #112845
by GiuseppeC

GiuseppeC replied the topic: DECIDE THE GROUPS ORDER

Dear holch thank you again for your reply.

I wrote only the conceptual problem in an easy way.

The reason why I need a solution is that I'm trying to implement a products rotation with LIME, and the groups of questions will be more than 2.

The correct example is this:

I want consumers to evaluate 3 (but maybe more) products. Every product has the same attributes (questions) to be answered to.

So for example I have

TAGLIATELLE

SPAGHETTI

RIGATONI

I make 3 identical GROUPS OF QUESTIONS (one for each product) and I have a presentation order, so every respondent has to taste the products in a given order.

As you say the fastest method (and really easy) is the Excel fixing solution. But suppose (only suppose) my boss told me that in this way there might be errors (I agree that filtering is a safe method )

The solutions I thought about are:

1) That one I told you about (JAVASCRIPT for changing destination in the result file)

2) conditions relate to an answer that is AFTER the filtering question (I don't know if it's possible)

3) A filter question that remand to other surveys (one for each product) with Expression Manager

What do you think about?

many thanks

I wrote only the conceptual problem in an easy way.

The reason why I need a solution is that I'm trying to implement a products rotation with LIME, and the groups of questions will be more than 2.

The correct example is this:

I want consumers to evaluate 3 (but maybe more) products. Every product has the same attributes (questions) to be answered to.

So for example I have

TAGLIATELLE

SPAGHETTI

RIGATONI

I make 3 identical GROUPS OF QUESTIONS (one for each product) and I have a presentation order, so every respondent has to taste the products in a given order.

As you say the fastest method (and really easy) is the Excel fixing solution. But suppose (only suppose) my boss told me that in this way there might be errors (I agree that filtering is a safe method )

The solutions I thought about are:

1) That one I told you about (JAVASCRIPT for changing destination in the result file)

2) conditions relate to an answer that is AFTER the filtering question (I don't know if it's possible)

3) A filter question that remand to other surveys (one for each product) with Expression Manager

What do you think about?

many thanks

Please Log in to join the conversation.

3 years 3 weeks ago #112848
by holch

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

holch replied the topic: DECIDE THE GROUPS ORDER

OK, if you have more options things can get a little bit more complicated (but I still think the easiest and safest solution is to let them all in their little space and then arrange everything later in the database, but I know how bosses can be...).

So you have 3 different products, lets say A, B and C. I think there still no other way than to have the different question groups for each product various times in the questionnaire. Maybe Tpartner has a Javascript solution for this? I am often surprised what he comes up with. But at the moment I can't think of anything else.

So there are the following combinations (if I am correct):

1. ABC

2. ACB

3. BCA

4. BAC

5. CAB

6. CBA

Hmmm, in this case even the order of groups in Limesurvey get's quite complicated... need to make a plan.

Regarding your solutions:

1. Don't quite understand it. What you you mean with changing the destination in the results file?

2. Also don't quite understand it. How can conditions help you here?

3. That doesn't help at all because it requires that you solve the problem later - in Excel...

Only thing that I can imagine at the moment is having an equation question for each actual question.

Lets make a simple example. You have Product A and B and you have Q1 and Q2 for each product.

So you have the following structure of the survey:

Group A --> Q-a1, Q-a2

Group B --> Q-b1, Q-b2

Group AA --> Q-aa1, Q-aa2

After that you put an equation question E-a1 and e-a2 and you write the results (via equation) from Q-a1 or Q-aa1 into this equation E-a1.

So you have 3 different products, lets say A, B and C. I think there still no other way than to have the different question groups for each product various times in the questionnaire. Maybe Tpartner has a Javascript solution for this? I am often surprised what he comes up with. But at the moment I can't think of anything else.

So there are the following combinations (if I am correct):

1. ABC

2. ACB

3. BCA

4. BAC

5. CAB

6. CBA

Hmmm, in this case even the order of groups in Limesurvey get's quite complicated... need to make a plan.

Regarding your solutions:

1. Don't quite understand it. What you you mean with changing the destination in the results file?

2. Also don't quite understand it. How can conditions help you here?

3. That doesn't help at all because it requires that you solve the problem later - in Excel...

Only thing that I can imagine at the moment is having an equation question for each actual question.

Lets make a simple example. You have Product A and B and you have Q1 and Q2 for each product.

So you have the following structure of the survey:

Group A --> Q-a1, Q-a2

Group B --> Q-b1, Q-b2

Group AA --> Q-aa1, Q-aa2

After that you put an equation question E-a1 and e-a2 and you write the results (via equation) from Q-a1 or Q-aa1 into this equation E-a1.

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

The following user(s) said Thank You: GiuseppeC

Please Log in to join the conversation.

3 years 2 weeks ago #113025
by GiuseppeC

GiuseppeC replied the topic: DECIDE THE GROUPS ORDER

I've done it :

I created a survey structured as follows:

CHOICE OF THE FIRST PRODUCT

CHOICE-1: Question 1 array with multiple response maximum allowable

1- TORTELLINI

2- RIGATONI

.....

FIRST PRODUCT GROUP

Q1_1

Q2_1

Qx_1

CHOICE OF THE SECOND PRODUCT

CHOICE-2: as the Q1 but with an exclusion filter array relative to Q1

SECOND PRODUCT GROUP

Q1_2

Q2_2

qx_2

CHOICE OF THE N PRODUCT

CHOICE-N : like the Q1 but with exclusion filter array on the Q ( N-1)

PRODUCT GROUP N

Q1_N

Q2_N

Qx_N

At the end :

RESULTS TORTELLINI

Q1_TORTELLINI : equation question that search in which group TORTELLINI were evaluated and SHOW THE RESULT CORRESPONDENT IN THE RIGHT ORDER ( the exact Q1_N )

ETC ...

equation questions will of course be hidden from consumers, but will appear in an orderly manner, in the data file .

The difficulty was in the fact that LIME has problems in equation questions related to multiple question (because they need subgroups ) so I had to use the formula IF((CHOICE-1_1.shown == TORTELLINI), Q1_1, IF ((CHOICE-2_1.shown == TORTELLINI ), Q2_1 , ........

THANK YOU AGAIN AND I HOPE THAT IT WAS CLEAR

I created a survey structured as follows:

CHOICE OF THE FIRST PRODUCT

CHOICE-1: Question 1 array with multiple response maximum allowable

1- TORTELLINI

2- RIGATONI

.....

FIRST PRODUCT GROUP

Q1_1

Q2_1

Qx_1

CHOICE OF THE SECOND PRODUCT

CHOICE-2: as the Q1 but with an exclusion filter array relative to Q1

SECOND PRODUCT GROUP

Q1_2

Q2_2

qx_2

CHOICE OF THE N PRODUCT

CHOICE-N : like the Q1 but with exclusion filter array on the Q ( N-1)

PRODUCT GROUP N

Q1_N

Q2_N

Qx_N

At the end :

RESULTS TORTELLINI

Q1_TORTELLINI : equation question that search in which group TORTELLINI were evaluated and SHOW THE RESULT CORRESPONDENT IN THE RIGHT ORDER ( the exact Q1_N )

ETC ...

equation questions will of course be hidden from consumers, but will appear in an orderly manner, in the data file .

The difficulty was in the fact that LIME has problems in equation questions related to multiple question (because they need subgroups ) so I had to use the formula IF((CHOICE-1_1.shown == TORTELLINI), Q1_1, IF ((CHOICE-2_1.shown == TORTELLINI ), Q2_1 , ........

THANK YOU AGAIN AND I HOPE THAT IT WAS CLEAR

Please Log in to join the conversation.