Sawtooth Software Forum - Recent questions

SSI Web v8.2.0 - download link

Sat, 18 May 2013 07:34:12 +0000

Dear Sawtooth team,

I instaled the updated version 8.2.2 but I face troubles with manual upload. So I want to return on v8.2.0 .

Could you please send me the link in order to download v8.2.0

Thank you in advance!

Maria

Export XLF without HTML, SSI Script, Javascript etc.

Fri, 17 May 2013 15:08:37 +0000

Can you export to XLF without all the script tags?

How do you add a graphic image (jpg) so it is on every page of the ACBC tournament?

Thu, 16 May 2013 20:33:59 +0000

Final utility versus calibrated utility

Thu, 16 May 2013 19:45:49 +0000

I used an ACA conjoint analysis for my master thesis. One of the hypotheses i want to test:

"Brand preference has a positive effect on the willingness to participate in a crowdsourcing contest initiated by that brand"

'Brand' is one of the in total four attributes (product, task, reward) that are included in my study.

I was planning to test if there is a correlation between the final utilities (brand preference) and the calibrated utilities (willingness to participate), but i am not sure if this is correct method?

Counting analysis CBC

Thu, 16 May 2013 15:01:23 +0000

Hi

In your CBC technical paper i read:

"Counting analysis calculates a proportion for each level, based on how many times a concept including that level is chosen"

Would it be wrong to say that the proportions could be biased since whether or not a concept is chosen depends on ALL the attribute (levels) in the concept? It could be that a level is often chosen, but it was actually chosen because of an other attribute?

If so could i say that MBC counting analysis is less biased in predicting price demand curves since it is an item specific and not concept specific analysis?

Best
Sandro

Select type on the same page

Wed, 15 May 2013 09:07:37 +0000

Hi all,

Would you have suggestions on how I could accomplish this problem –

I have a grid-type question with 3 rows and 4 columns. Each of these 4 columns have a drop-down list of answers. So in total there are 12 dropdown lists to answer. However, I would like Column 2 list to appear only if a specific option in Column 1 is selected, and this should still be in the same page of the survey.

Is there a way to work this out, without having to submit first the page before this take effect?

Appreciate your help! Thanks

1 member constructed list and automatic skipping

Tue, 14 May 2013 15:51:16 +0000

I have a constructed list utilizing unverified perl that is functioning properly. I have tested to ensure it returns members effectively and the logic is right.

When only one member is returned in the list based on the logic it skips over the next question in the survey where these members of the list are supposed to be piped into. If I have 2 or more members in the list it asks the question fine.

If there a way to get around this?

Is it possible to add headings to the CBC/HB draw file?

Tue, 14 May 2013 15:51:05 +0000

Hello!

When I run the CBC/HB module and save a random draw file, I notice there are no headings of attribute levels in the draw file. Is there a way to save the random draws with headings?

Thanks!

Batch MaxDiff scores computation in SSI Web

Tue, 14 May 2013 14:24:24 +0000

Hello!

We are currently having a 16-country project with MaxDiff.
All the countries use the same questionnaire template and all the 16 MaxDiff designs are identical (using the same design file).

Soon I'll have to compute the scores, and it is to be done on a country level, so I started thinking about ways to automate the scores computation process.

I'd like to run the computation in SSI Web in order to not deal with HB utilities rescaling.

Could you please tell me if there's a way to program a batch estimation in SSI Web? Does SSI Web have an API library at public disposal?

I've tried to solve it using AutoIt script, but it seems impossible to interact with SSI Web controls, and now I see the only way to program it as a series of mouse clicks on the main window.

Could anyone offer a smarter way, I would highly appreciate that!

Thanks a lot,
Igor Gusev

CBC price attribute - Continuous Price

Mon, 13 May 2013 19:04:08 +0000

Hello

We are looking at a CBC design with 10 SKUs and 5 price levels which we will show as conditional pricing (very high / high / medium / low / very low).

What we would like to do is offer the client the possibility to simulate any price he desires. For instance, if sku A had the price levels set at $5 / $6 / $7 / $8 / $9 we would like to be able to run simulation with prices such as $6.49 or $7.35.

I read Bryan Orme's paper "Three Ways to Treat Overall Price in Conjoint Analysis" and it seemed to indicate that it is possible through the continuous price approach, however it's not fully clear how to use it.

So my questions are:
1. Is this possible?
2. If yes, how? Should we edit the .cho file and if so in which way?

Thank you for your help inadvance.

Importing CiW data into Stata

Mon, 13 May 2013 15:22:56 +0000

Looking for tips on the fastest, most efficient way to bring CiW data into Stata. Is Import in fixed format with dictionary the best way to go? any issues to watch out for?

Skip logic on grid question

Mon, 13 May 2013 07:55:39 +0000

I have a problem to set up a skip logic based on answers to a grid question.
I would like a page to be skipped if the respondent does check the answer on row 3 and column 5 on my grid question, called "prepacked".
So in the Skip logic section "Skip if the following logic evaluates to true" of the second page I've put the script:
prepacked_r3_5 = 1 or prepacked_r3_c5 = 1
but when I checkd it I receive the message: "The sympbol 'prepacked_r3_5' does not exist"
Any clue on what I'm doing wrong?

Also, how can I check that my redirections are working properly before I upload the questionnaire?

Thanks in advance

How do you hide a CBC design that is used to populate Free Format questions?

Fri, 10 May 2013 18:05:04 +0000

Hello,

We tried placing the following comment in the header of the CBC exercise:


We get the following error when testing:
Sawtooth Error #147
Custom CBC error

When the comment is removed the free format question appears as it should, but so does the CBC exercise that the Free Format question is based on.

Thanks!

Request Link for identifying users from different countries

Thu, 09 May 2013 16:35:25 +0000

Hi there,

I received from the company three request links for different countries. We are sending survey for customers from three countries. Those links should identify users, thats why it has id for the country, user email and password.

It looks like this
http://news.company.com/automated/action.jsp?action=updateRecipientMsg&errorPage=//automated/action.jsp&pemail=XXXX@XXXX&gid=ID&uemail=
executor@news.company.com&psw=XXXX&self=true

The problem is, that I don't know where to put those links in sawthooth, cause normaly to request survey we use the link like it is described in Sawthooth Help
http://www.sawtoothsoftware.com/gonzo/cgi-bin/ciwweb.pl?studyname=tep1&mypassword=XXXXX

Does anybody know how to handle it ?
Thanks in Advance.

Kind Regards
Rita

how can ı do copy- paste

Thu, 09 May 2013 05:24:10 +0000

yes i have survey data in Excel and i want to construct a logit model using this result for this reason i want to use sawtooth, but excel is .xls file how can i open this file in the sawtooth. i want to estimate beta value for my attributes.

Self Hosting Sawtooth Error #117

Wed, 08 May 2013 17:05:02 +0000

Hi, I've trying to upload a survey (made in Sawtooth 8.2.2) to a cPanel host. The web server management report keeps showing me the message "Sawtooth Error #117 File open error" It appears when the command status is "Checking database for survey name conflicts...failed". It seems to be related to perl files, since the connection status stops when "-rwxr-xr-x admin.pl" and "-rw-r--r-- authlib8_2_2.pl". Is it because I have to specify the path to perl modules in the server? I'be trying to change the scripts of admin and ciwweb, but it doesn't seem possible. Thanks for your help.

CBC design avoiding no brainers

Wed, 08 May 2013 16:32:09 +0000

I'm setting up a CBC with 5 attributes all or which are clearly ordered i.e. the levels are best to worst. This means some of the choices are clearly no-brainers i.e. one option is better on all attributes compared to another. Is there any way of avoiding these in the design?

Thanks All

how can i import to excel file to sawtooth

Wed, 08 May 2013 12:22:41 +0000

i made a internet based survey so i have result of this survey and i want to use sawtooth but i didnt find how can import to sawtooth ? is it posibble ?

Function for getting Max Value in Ranking

Wed, 08 May 2013 08:16:55 +0000

I'm building a custom script for a ranking question, and would like to access the Max Value set in the options for the question.

I'm looking for something similar to ErrMax() but for use within the question.

Using javascript to compute - over a LOT of variables

Tue, 07 May 2013 23:55:39 +0000

Hi all ...

Here's the problem: a heap of codes representing places (so we have a start point of a journey and an end point). We then work out the distance travelled. We take the time that the respondent said the journey took and the mode of transport they say they used. We then use this information to ascertain whether the journey as reported is physically possible. If not, we ask the interviewer (this will be an interviewer-assisted survey) to confirm the inputs with the respondent. Once all inputs are confirmed, the interviewer can then click a code and continue (even if the respondent doesn't want to change anything - at least we have it flagged as checked). The respondent can have multiple journeys and there are a heap of places they can travel between.

My first thought was that I'd need to build this in perl - but of course perl code of this magnitude will slow the survey to a crawl because there would need to be a lot of checking back with the server. But if a table were built using javascript, it shouldn't slow things down much at all.

Problem is that I don't know enough javascript to hang myself with. I just know enough for simple validations etc.

Has anyone done anything like this before? And can you point me in the right direction? I'm more than happy to pay the Sawtooth guys to write it for me if necessary, but it'd be great if I could learn something new and do it myself ...

Thanks in advance. Oh - we're on 8.2.2, and the survey will be on tablets using mobile broadband.

Link two parts of a survey checking quotas

Mon, 06 May 2013 16:53:29 +0000

Hi there,

I have a big questionnaire that needs to be slipt in two parts.

If we programmed two separate surveys, we would need to pass a lot of info between the two surveys, will have to replicate a lot of lists which are used in both the parts and will also have to match the two datafiles.

To avoid all these, we would like to programm a unique long questionnaire, interrupting the completion using a Link question at the end of the first part and redirecting people to a Thankyou page which reminds them that they will be able to take the second part of the survey after a couple of days.
Later we would send respondents a link which allows them to re-start the survey from the beginning of Part2.

All this works properly but the problem is that we should also be able to control quotas based on the number of completes for Part1.

Any idea on how to manage this?
(hope I was clear in my explaination)

Thanks,

Alessia

CBC design in multiple languages

Mon, 06 May 2013 16:03:08 +0000

Hi Sawtooth experts,

I have a study with multiple languages. Is it imperative that the CBC design that I generate is only 1 across all these language versions? That is to say, I generate a design for the English version (number of qnaire versions=300, design seed=1, all other aspects are the same - number of tasks, items, etc.). Then I just translate the other questions into different languages, without re-generating another design. At the end, I would have to combine the results for these. What would be the statistically best action for this? Thanks

How can MBC and CBC results be compared?

Mon, 06 May 2013 09:30:18 +0000

Dear Experts

In order to answer our research question we have conducted two independent studies. One time using the MBC method, the other time using CBC. The data of both studies have been analyzed through HB estimation. Subsequent market simulations produced contradicting results (in order of prices and product bundles)

Are there any scientific approaches to analyze the contradicting results? Also can you recommend any papers that cover the impact of different approaches of data collection on the results?

Best
Sandro

how do you artifically creat a .dat file?

Sun, 05 May 2013 22:32:23 +0000

what is the formular of calculating average utility

Sat, 04 May 2013 15:13:48 +0000

Merge MaxDiff data

Fri, 03 May 2013 17:06:13 +0000

Hello all,

I have collected data on an online MaxDiff survey. When trying to update the survey, I accidentally erased the data already collected from the admin module. However, I had previously downloaded the respective .dat file. Now that I have more data collected, I want to merge these two sets of information to compute the scores. Both sets have the same design file, so my question is: how can I do it?

(I already saw both http://www.sawtoothsoftware.com/forum/3322/merging-similar-maxdiff-studies and http://www.sawtoothsoftware.com/forum/3057/how-to-merge-data-for-a-maxdiff, but it seems they are somewhat different from what I'm asking)

Many thanks.

How to use "Quota Control Functions"

Fri, 03 May 2013 13:04:56 +0000

Hi,

Just wondering how to use Value(QuotaFromQuestion) function ...

Did some tests but couldn't get any value other than -2 .

Thanks

(Version 8.2.2)

How can I assign a random subsample to take one portion of a survey, and the other to take another portion?

Thu, 02 May 2013 19:51:54 +0000

I'm programming a study where I will have 3,600 respondents overall. I want 400 of these to do a conjoint, the other 3,200 to take the regular survey, and all 3,600 to answer a demographic section at the end. Rather than just performing a separate conjoint study (concerned about costs), is there a way to do this and keep it at one study?

Instead of skipping to and exercise can somehow skip and exercise completely?

Thu, 02 May 2013 14:32:17 +0000

Out of Memory

Wed, 01 May 2013 17:29:01 +0000

Ever since upgrading to SSI Web 8.2.2 I experience the error "out of memory" when I try to export data into an .sav file. My program uses 77 data fields and I am only trying to export 4 test cases. I have already tried exporting only half of the variables and I still received the error message. Does anyone have any advice as to what I check next to make sure this isn't happening tomorrow when the survey goes live?

Lookup lists - or dynamically reducing list

Tue, 30 Apr 2013 21:42:46 +0000

Hi all ... I have a large interviewer-assisted project coming up, and have been asked about a particular sort of functionality. Obviously not do-able "out of the box" (would need some sort of javascript/etc) but keen to find out if anyone has successfully implemented this.

Here's the picture:

Say we have a really big list. Maybe a list of all the locations/small towns/suburbs in a state. The interviewer needs to understand which exact place the person last visited, but (a) the person might not remember straight away - they know it was somewhere with a name starting with N ("Narra-something") - and (b) the list is really long anyway. SO ... the interviewer (faced with a really big list) just enters the letter N. The really big list starts to reduce - now only showing the places beginning with N. S/he then enters A. Now we're getting somewhere. The list now reduces to places beginning with NA. We see Narrabri and Narrabeen in the list.

What do you reckon? (p.s. on 8.2.2)

Closing browser window

Tue, 30 Apr 2013 20:42:34 +0000

On the closing screen my client wanted a link to close the browser (instead of just saying "you may now close your browser").

I've tried a couple of things below and they worked fine in Internet Explorer, but don't do anything in Mozilla. Any ideas?

Tried this button:
<input type="button" value="Close Browser" onclick="window.close()">

Tried this link:
<a href="JavaScript:window.close()">Close Browser</a>

How do you skip a participant to another BYO set based on an specific answer to a initial question?

Tue, 30 Apr 2013 20:22:22 +0000

I have a ACBC that I want to direct participants to another byo that has different price attributes if they answer a certain response to an opening question.

AIC Function Feature in Constructed List

Tue, 30 Apr 2013 18:48:35 +0000

Hello,

I have two multiple select question.

The first question contains the below specified attributes and ask respondents choose aware one.

Q1. Which of the following online email services that you aware?
1. WindowsLive/Outlook
2. Gmail
3. Yahoo Mail!
4. Others, Specify
5. No Idea

The second question contains same attributes with change in last attribute. Here we are asking respondents based on the selection of previous questions which are all you are using?

Q2. Which of the following online email services you are using?
1. WindowsLive/Outlook
2. Gmail
3. Yahoo Mail!
4. Others, Specify
5. Not using anything

So, I have created constructed below. It is working fine.

AIC(Q1)
REMOVE(Q1List,5)

However, in Q2 I want to show the attributes selected in Q1 + Not using anything.

How to achieve this?

Drop down option in a grid type question

Tue, 30 Apr 2013 15:10:50 +0000

Hi, I have a grid type question in which I would like the colums to have a drop down list of items.
For rows I have
Item 1
Item 2
Item 3

And for columns, I have 3 sets of choices (3 columns)
1st set - item 1, 2, 3,4,5,6
2nd set - item 10,11,12,13
3rd set - item 21,22,23,24

I want this available on each item on the row as combo box/dropdown menu.

Any ideas as to how I could do this? Thanks!

Web surveys on tablets

Tue, 30 Apr 2013 03:15:48 +0000

Hi all

Got a BIG potential project coming up where we will be doing central-location, interviewer assisted surveys. They will all be in places where there is good wi-fi, so I am thinking that we'll set it all up as a normal web survey. That way we can monitor progress remotely, and if the tablets are lost or stolen we haven't lost the most valuable bit - the data.

Does anyone have any first-hand experience of this? What brand of tablet would you recommend? Everyone thinks of iPads first but I'm trying to keep an open mind. Interviewing will be pretty continuous over at least three years, so we'd want something pretty bullet-proof. I'm also thinking that having them SIM enabled would also give us a secondary way of connecting if the wi-fi in an area goes offline.

What do you reckon?

2013 Sawtooth Software Conference Announcement

Mon, 29 Apr 2013 14:21:03 +0000

The 2013 Sawtooth Software Conference will be held October 14-18, at the Laguna Cliffs Marriott Hotel in Dana Point, California, USA. This conference is held only once every 18 months. Given this location and the success of our last conference, we expect 275 to 325 people to attend.

Registration is not yet available for the conference; we are just giving you advanced heads-up so you can see the quality of the program and put it on your calendar as a must-attend event for this fall.

During the first two days, October 14-15, we're offering optional 2-day workshops and 4-hour tutorials. There will be three 2-day workshops (CBC, MBC, and a CBC Consulting Challenge). The 2-day workshops will cost $1,100.

The cost of attending the main sessions on October 16-18 is $1,250 (if payment received by August 23).

The main sessions consist of approximately 30-minute papers followed by audience Q&A. Read more details, including presentation abstracts at http://www.sawtoothsoftware.com/education/conf13.shtml

Below are the paper titles and authors for the short presentations delivered over Ocrober 16-18.

Wednesday, October 16:

"Nine Things Clients Get Wrong about Conjoint Analysis"— Christopher Chapman, Google

"Quantitative Marketing Research Solutions in a Traditional Manufacturing Firm: Update and Case Study"— Robert J. Goodwin, Lifetime Products, Inc.

"Can Conjoint Be Fun?: Improving Respondent Engagement in CBC Experiments"— Jane Tang and Andrew Grenville, Vision Critical

"Making Conjoint Mobile: Adapting Conjoint to the Mobile Phenomenon"— Chris Diener, Rajat Narang, Mohit Shant, Hem Chander, and Mukul Goyal, AbsolutData

"Choice Experiments in Mobile Web Environments"— Joseph White, Maritz Research

"Using Complex Choice Models to Drive Business Results"— Karen Fuller, HomeAway, Inc. and Karen Buros, Radius Global Market Research

"Augmenting Discrete Choice Data – A Q-sort Case Study"— Brent Fuller, Mike Smith, and Matt Madden, The Modellers

"MaxDiff Augmentation: Effort vs. Impact"— Urszula Jones, and Jing Yeh, Millward Brown

"When U = βx Is Not Enough: Modeling Diminishing Returns among Correlated Conjoint Attributes"— Kevin Lattery, Maritz Research

"Respondent Heterogeneity, Version Effects or Scale?"— Keith Chrzan, Sawtooth Software

Thursday, October 17:

"Bridging Survey Research with Social Media Monitoring Services"— Karlan Witt and Deb Ploskonka, Cambia Information Group

"Brand Imagery Measurement: Assessment of Current Practice and a New Approach"— Paul Richard McCullough, MACRO Consulting, Inc.

"ACBC Revisited"— Christopher Fotenos, Jeroen Hardon, and Marco Hoogerbrugge, SKIM Group

"Research Space and Realistic Pricing in Shelf Conjoint"— Peter Kurz, TNS Infratest, Stefan Binner, bms marketing research + strategy, and Leonhard Kehl, Premium Choice Research & Consulting

"Attribute Non-Attendance in Discrete Choice Experiments"— Dan Yardley, Maritz Research

"Anchored Adaptive MaxDiff - Application in Continuous Concept Test"— Rosanna Mau, Jane Tang, LeAnn Helmrich and Maggie Cournoyer, Vision Critical

"How Important Are the Obvious Comparisons in CBC? The Impact of Removing Easy Conjoint Tasks"— Paul Johnson and Weston Hadlock, SSI

"Segmenting Choice and Non-Choice Data Simultaneously: A How to..."— Thomas C. Eagle, Eagle Analytics of California, Inc.

"Extending Cluster Ensemble Analysis via Semi-Supervised Learning"— Ewa Nowakowska, GfK Polonia and Joseph Retzer, MarketTools Inc.

"The Shapley Value in Marketing Research: 15 Years and Counting"— W. Michael Conklin and Stan Lipovetsky, GfK

"Demonstrating the Need and Value for a Multiobjective Product Search"— Scott Ferguson and Garrett Foster, North Carolina State University, and Joseph Donndelinger, General Motors Research and Development

"A Simulation Based Evaluation of the Properties of Anchored Max/Diff: Strengths, Limitations, and Recommendations for Practice"— Jake Lee, Maritz Research and Jeff Dotson, Brigham Young University

Friday, October 18:

"Contexts in Which Best-Worst CBC Are Most Valuable: Application to School Choice"— Namika Sagara and Joel Huber, Duke University, and Angelyn Fairchild, Research Triangle Institute

"Does the Analysis of MaxDiff Data Require Separate Scaling Factors?"— Jack Horne and Bob Rayner, Market Strategies International

"How to Use Conjoint to Determine the Market Value of Product Features"— Greg Allenby, The Ohio State University, Jeff Brazell, The Modellers, John Howell, The Ohio State University, and Peter Rossi, UCLA

"The Ballad of Best and Worst"— Tatiana L. Dyachenko, Rebecca Walker Naylor, and Greg M. Allenby, The Ohio State University

Second price function

Mon, 29 Apr 2013 13:42:37 +0000

Hello,
we want to add a second price function in an ACBC study.
This function should multiply the value of the summed-price function with X.
Is there an opportunity to add such a function and how?
Thank you for your help.

Getting the name in "Others"

Mon, 29 Apr 2013 09:21:02 +0000

Hi, I made a constructed list which include an option "Others,please specify" which is Item 5 in the parent list. Problem is, in the next question, when I use the constructed list, the same "Others,please specify" appears, and not the actual answer specified.

Below is the script I used for the constructed list. What do I need to modify from this to get the actual name from "others'? Thanks.

Begin Unverified Perl
my $i=1;
for($i=1;$i<=10; $i++)
{
if (VALUE("Q05BrandsubUsage_r".$i."_c1")==1)
{
ADD("Material",1);
}
if (VALUE("Q05BrandsubUsage_r".$i."_c2")==1)
{
ADD("Material",2);
}
if (VALUE("Q05BrandsubUsage_r".$i."_c3")==1)
{
ADD("Material",3);
}
if (VALUE("Q05BrandsubUsage_r".$i."_c4")==1)
{
ADD("Material",4);
}
if (VALUE("Q05BrandsubUsage_r".$i."_c5")==1)
{
ADD("Material",5);
}
}
End Unverified

How to re-weight CBC importance scores

Fri, 26 Apr 2013 21:05:28 +0000

If I want to get the importance scores from a CBC exercise, the number of levels across attributes should be the same. If not, the one with the more levels would end up becoming more important. Is this correct? If yes, is there a way to re-weight the attributes to deflate the impact of the attribute with the higher number of levels outside of conjoint?

Matching a numeric response to a list

Fri, 26 Apr 2013 20:40:39 +0000

I have a question (Q1) asking respondents for their 5 digit zipcode (I set this up as an Open Ended with some javascript to make sure it only contains numbers. This is to allow for zipcodes beginning with a 0).

I have a list with 265 zipcodes in it and I need to set a skip if their response at Q1 matches any of the zipcodes in that list. I'm a bit foggy on the easiest way to accomplish this. Any help?

Prohibit checking both columns in column-based grid

Fri, 26 Apr 2013 18:45:00 +0000

Hello,

I have the following question and I'm having trouble programming the grid:

Overall, which program feature is the MOST valuable to you as a customer and which is the LEAST valuable?
Feature 1
Feature 2
Feature 3
Feature 4
Feature 5
Feature 6 (other/spec)

Most Valuable
Least Valuable

I only have 6 features, so I can't do Max Diff...but it would basically need to operate like that.

My first instict was a radio column-based grid, where they could only check one item as Most Valuable and one item as Least valuable. The trouble is, this allows them to select one feature as both the Most AND Least valuable. Is there a way to prevent this?

I actually thought about programming it as a one-page MaxDiff exercise, but it would make the data a pain to analyze on what is otherwise a pretty simple survey.

Thoughts?

Computing Power for Big HB Runs?

Fri, 26 Apr 2013 15:57:07 +0000

Recently I have come across a couple datasets where I have tried running covariates, constraints, and (significant) interactions together (with main effects too, of course) and I end up with around 200 or so parameters. With 1200+ respondents, this is taking my computer 150-300 hours of computing time - 0.04 iterations per second).

My computer specs:
AMB A8-4500M APU Quad core - 1.90 GHz
5400 RPM 600 GB HDD
6 GB RAM
64-bit

I've discussed with our partners and we're willing to buy another computer dedicated for running Sawtooth, but we don't know what type of computer would be best. Is the bottleneck:

HDD vs. SSD?
CPU speed?
# of CPU cores?
RAM?
Graphics card?
Something else?

I am looking for information on if a new computer should help, and what specs are most important for this new computer?

Thanks!!

Trouble with data generator since update?

Fri, 26 Apr 2013 14:53:42 +0000

Just updated to 8.2.2 from 8.1 and now the data generator isn't working. No matter how many respondents I tell it to generate, it only generates 1. Is anybody else having this problem?

Video Streaming in Google Chrome

Thu, 25 Apr 2013 20:37:02 +0000

I have some inconveniences with Google Chrome when trying to show an Ad in video included in a survey from a web server. When I use Internet Explorer it works fine. Is there any way/method to be able to use Google Chrome too?

Calculating the Monetary Value of 36 Attributes

Thu, 25 Apr 2013 17:58:45 +0000

Hi,
We have a client that currently offers its customers 5 product categories with varying benefits. They have a list of 36 benefits (with 2 to 6 levels) that make up the 5 products. They wish to simplify their product offering by removing some of these benefits from their list. To do this, they wish to know the monetary value of each of these benefits.
We were thinking of using Partial-Profile CBC, or some type of Max-Diff. If anyone has any ideas about how this can be done most effectively, your opinion would be greatly appreciated.
Thank you very much,

Shahrokh

How to force show a question?

Thu, 25 Apr 2013 10:29:27 +0000

A question of select type is getting skipped because it has a single response. But I want to show that page. How can I do that?
Please help.

Conditional relationship in grids

Thu, 25 Apr 2013 07:40:05 +0000

Hey guys,

I am trying to get a conditional relationship between two grids.

The first grid consists of 3 columns displays product labels (images) on the first column, the second column asks to check box if this label is known by the respondent and the third if the label is not known to them.

Now I want to pipe all the labels that are known (checked boxes in second column) into my next question, which is also a grid with the same pattern (i.e. 3 columns) but now displaying in the first column only those labels that were in the previous question indicated as known.

In the second column it is then asked whether the respondent thinks this label is related to environmental friendliness (check boxes) and in the third whether s/he thinks it is not (check box).

Any ideas on how to do this?

Thanks a lot in advance.

Max Diff Analysis excluding some of the statements evaluated

Thu, 25 Apr 2013 07:06:45 +0000

Hi all,

We are about to design a customer satisfaction survey and plan to use Max Diff approach to determine the ranking of dimensions.
We were thinking also of having in the max diff exercise statements covering other touchpoints than the ones evaluated through the q're, eg. brand. Questions is: if it's feasible technically/make sense to do the max diff analysis excluding some of the statements.

Any of your thoughts on this, much appreciated!

Marin

Script on degree of familiarity

Wed, 24 Apr 2013 15:44:58 +0000

Hi Sawtooth experts,

Would you have recommendation on what I could use (script maybe) for this problem:

I have a question on familiarity of 3 products (5-level scale, not familiar to extremely familiar). Now a respondent will only have to rate 1 of these 3 products. If he/she is aware of all 3 products, the selection will be the degree of familiarity. The product in which he/she is most familiar with will be rated.

I thought of creating a tag for the respondent to lessen the script for the logic, but couldn’t think of a way to incorporate this. Hope you could help. Thanks a lot!