iQ Suite

Hi, Is there a way to analyze MaxDiff data outside of the built-in Qualtrics program? Or can I import my MaxDiff data into qualtrics to run a MaxDiff analysis? Context: I ran a MaxDiff survey and had to republish it midway through data collection, and unfortunately the only way I found to do that (survey approver survey) was to make a tiny change to a feature in the MaxDiff and then to change it back. This unfortunately reset the MaxDiff analysis, and now I can only run the analysis with new data collected since the republishing. I tried importing the dataset into a copied survey, but the data doesn't load into the MaxDiff like it normally does. Any ideas? Is there another way to analyze it? Or reset the maxdiff analysis program somehow?

This support page states “Once a MaxDiff analysis is reset, the previous data is irretrievable.”Does this mean that all previous data is deleted or merely not accessible on the MaxDiff Analysis page? Does “previous data” include both individual respondent maxdiff utility scores and data from the rest of the survey?I have a maxdiff that's running live and has already collected data. I have some aesthetic changes that need to be made to some of the maxdiff items -- however, Qualtrics warns against making edits to a live maxdiff because (A) it could influence respondent choices in the maxdiff and (B) it will reset the MaxDiff Analysis page. However, I'm not worried about A and my team is using Tableau to analyze the data so I don't need B. Instead I merely need the individual respondent utility scores. So -- does anyone know more specifically what the impact of editing a live maxdiff is in terms of what happens to the previous data? I basically just need to know wh

Hello, I have five questions in my survey that fall in one subscale. For example, 5 questions fall under a 'great heights' subscale. Respondents will answer each question on a Likert Scale (strong yes -> strong no) individually. I want to see the combined response count on the Likert scale for all 5 questions. How can I combine the responses from the 5 questions to receive one described data (mean, response count, etc.).I hope this makes sense.

We have a lot of responses to our quick feedback form that are short phrases eg 'it's great' or 'I hate it' but don't include other contextual phrases or key words to help us categorise.Is it possible to do a search string in TextIQ for everything under, say, 7 words? (Then we could exclude other keywords that would live in our other topics.

I understand how to create the Word Cloud visualization from my open text survey responses. However, once I've created the Word Cloud, I am looking for the word counts that have to be tabulated on the back end to generate it. The Word Cloud is the visualization of the counts...but where can I find the data of the frequency counts?

Hi everyone,I created a survey, where half of the respondents got to see a specific message, and the other half got to see another one. The questions following after this message were the same for everyone, only the message was manipulated for the two groups. Now I am looking for a way to analyze those answers, and compare the answers of the group that saw message A versus the group that saw message B. Does anyone know how I can manage to do this? Much appreciated!

I'd like to include analysis by system start date but it doesn't show up as a choice in my crosstabs or as a variable for bucketing. How do I include the system start date in my analysis?

I have a Pick, Group and Rank question in my survey. I am trying to make meaningful analyses of the question using statsiQ. Can anyone help with it?

Hello,In Design XM solution;- With MaxDiff - TURF analysis: what is the difference between the calculation types: first choice and top 2 choices?- With Conjoint report: what is the difference between the calculation types: preference share, first choice, utility scores, cost analysis?Thanks for you help😊

Firstly, apologies for not posting on the Stats iQ board. It's currently being inundated with spam (7 pages of it) and this is a relatively urgent query, so I don't want it to get buried. My company has a site licence for Qualtrics. One of our users has reported that the Crosstabs option has disappeared from the Data & Analysis tab. Has this been moved elsewhere/is it now part of the Stats iQ addon?

Como agrupar dos preguntas con diferentes tipos de datos y mostrar los resultados en un Dashboard?

I currently have segment values assigned to customers, and a calculated "Segment Movement" numerical value that shows both the direction and the magnitude of the change between loyalty values on a scale of -4 to +4. However, I have several experience ratings (input variables) for a respondent to select from 1 to 5. Is there a way to account for the fact that the output variable includes both negative and positive values in regression in Stats iQ?

Dear all,I'm seeking to correlate different key drivers to a NPS.The key drivers exist both in text set (not satisfied to fully satisfied) and in number set (1 for not satisfied and 5 for fully satisfied).When I use regression analysis from StatsIQ with the NPS as the key variable, the results are different depending if I correlate to the text set or the number set. If I do the correlation using text set drivers, let's say that drivers #1 and #2 will be the most correlated to the NPS. But if I do the correlation using number set drivers, drivers #3 and #4 will be the most correlated to the NPS.Thus, I'm not too sure which analysis is the one I should rely on to draw conclusions. Have you already faced this question? Would you be able to share some insights on how to interpret the results?Many thanks in advance.FJ

I have randomization in my survey to create two experimental conditions. How do I get the information about this randomization to show up as a variable when I am in crosstabs?

Hello everyone. Is there a way to automate exports from crosstabs?

Hello! Please help a newbie!Can anyone advise how I can take a Likert scale data (strongly agree, agree, neutral, disagree, strongly disagree) and group the (strongly agree/agree)/(strongly disagree/disagree) totals, and then pull this into a report text area to say "XXX% of respondents either agreed or strongly agreed to the statement xxxxx"?I cannot figure out how to correctly bucket the data and then take this variable into the report. Is it in "Data" tab or is it in Stats IQ tab? and is it possible to take the variable and insert into a report?I suppose the same goes for CrossTabs - can these charts/pivot tables etc be used in the reports? and how do i get them inserted?Eventyally what I need to happen is have one overall report that I detail some visualisations, then sum up with some text i.e. XXX% of respondents either agreed or strongly agreed to the statement xxxxx, where as XX% disagreed or strongly disagreed. I will then filter the report for the

Hello, I would like to create a survey on the XM platform that enables participants to prioritize a large number of features, say 20-25.I have read that MaxDiff is a good survey technique for this. However, only the PX platform supports full MaxDiff projects. We do not have the budget or the time it would take to upgrade to PX from XM…that may be in the future but not now. In the meantime, what are work arounds and options? Pros and cons of them?Qualtrics support has told me I could program a real MaxDiff on the XM platform but it would be very difficult. Your thoughts?Or, how could I develop a series of questions on the XM platform that uses some MaxDiff principles, even if not a full-blown MaxDiff? If so, what would it be?My fallback option is to use the multiple question type to ask 2 rounds of questions. The first round is a series of questions so that every feature is seen once...each of the 4 questions would have 6 features as answers (all randomi

Hi! I have performed a conjoint analysis. Automatically the simulator performs a Bayesian estimation. However, I want to perform an MNL estimation by myself. In order to do so, I need the complete information about all conjoint questions: the actual raw data. This seems not to be possible. Does anyone know how to do this?Regards,Tessel

We are a group of researchers and StatsIQ is very helpful to us specially with our colleagues in the academe. We just want to know how does Qualtrics StatsIQ know that a certain test is applicable or recommended? For us research professionals we can certainly know what test to do. However, we have encountered a test that StatsIQ recommended an ANOVA, however the test should be Chi-Squared Test. This will help us understand how StatsIQ runs the recommended test so we can share it to our colleagues.

Is there a way to filter comments by sentiment in Text IQ or do you have to go into the data & analysis or report? It would be helpful to filter in Text IQ.When I change the sentiment assigned to a comment or a topic, how do I save my changes?I'd also like to generate a bubble chart that illustrates the topics, but for my particular survey, the sentiment analysis is not relevant. Can I depict the topic bubbles without the sentiment?

I have two questions that are related to one another. Based on the answers selected in both questions it will determine whether the third question will be displayed or not. Is there a way that I can combine both questions and place a condition that will determine whether the third question should be displayed or not?

I can not find any details about the methodology/modeling used for the anchoring of Max-diff. Below is all I could find so far, but it's not much help. Any additional insights?Anchored MaxDiff WHAT IS ANCHORED MAXDIFF? Anchored MaxDiff is a supplemental methodology where a follow-up question is asked after each MaxDiff task. It has some similarities to dual-choice conjoint analysis both in how the question is asked as well as how it is modeled. The approach includes asking a question immediately after each MaxDiff task. After being presented with the list of items, the respondent is asked whether: All of the items they see above are important/preferred. Some of the items they see above are important/preferred and some unimportant/not preferred. All of the items they see above are unimportant/not preferred. This data is factored into the statistical model. It yields an understanding of the anchor point for where the utility output is above and below a line where it

I am testing my survey before full launch and I see 14 responses in the "reports: survey results" tab, but when I click on the "reports: conjoint analysis" tab it tells me it can not calculate because it needs 4 responses and only has two. I was testing it in "preview mode". Should I be worried responses are not being counted? (Most of them are mine from testing) I am hesitating fully launching this survey - obviously, I want the conjoint responses counted if respondents completed that question block. Any ideas why this is happening? Should I be worried?

Hi Experts, I'm facing an issue in Stats iQ where I'm unable to view any data for analyzing. I've imported 3k historical data responses and I've not created any filters yet, I tried using Describe, Relate etc to analyze the data and I'm unable to view any info, below are the screenshots for reference. Appreciate if anyone can help me here. ! ! ! Regards Pradyp

We have a survey we deploy every six months. We download the survey data as a csv to do our analysis using statistical software. The download currently works with the legacy exporter, but does not work with the standard exporter. Customer support indicated that the reason is that our survey is “too long.” I find this surprising. There are approximately 58,000 columns in the exported file due to the looping structure of the survey (with fewer than 20 questions), and the final file size is less than 30 MB, which does not seem unreasonable for a csv file. Customer support tried to download the data using the standard exporter and wasn't able to do it (only the metadata is exported). They suggest reformatting our survey but the solutions they suggest either won’t work for us (if we allow write-in answers and pipe those forward instead of using the dropdown selections that our loops are based on, we’ll have to spend a significant amount of time cleaning and re-categorizing thos