Stats iQ | XM Community
Skip to main content

iQ Suite

Discuss Stats iQ, Crosstabs, and Imported Data Project solutions, best practices, & more.

  • 203 Topics
  • 402 Replies
203 Topics
kai_scott
kai_scott
asked in iQ Suite
Relating checkboxes and categories variables

I am a big fan of Stat iQ, but have run into some issues when relating a variable with check-all-that-apply (e.g., types of discrimination) and a category (e.g., gender). The kicker is that the responses to the types of discrimination is based on display logic of those who responded "yes" to the previous question asking whether or not they have experienced discrimination in the workplace. So, not everyone who responded to the gender question is responding to the types of discrimination question. When I export the results after relating them in Stats iQ, I get a lot of wonderful graphs, some of which are based on a sample size for all who responded to gender and another sample size based on all the boxes checked for each response option. But there doesn't appear to be an option to relate to the response count of the types of discrimination question for each gender category. I ran the numbers manually (using filters in the results section) and came up with very different results, which

640
kai_scott
2 years ago
P
PriyaGupta
asked in iQ Suite
How can we best analyse the Pick, Group and Rank question in statsiQ?

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?

2021
KatF
Qualtrics Employee
2 years ago
R
Rubee
asked in iQ Suite
The meaning of calculation types in Design XM reports

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😊

291
KatF
Qualtrics Employee
2 years ago
D
DanRyan
asked in iQ Suite
Crosstabs disappeared?

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?

5595
D
2 years ago
C
CarlosCG
asked in iQ Suite
Variables o Agrupaciones de Stats a Dashboard

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

61
AmaraW
Former XM Community Team Member
2 years ago
S
southboundphoenix
posted in iQ Suite
Regression With Both Positive and Negative Values in Outcome Variable

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?

440
S
2 years ago
F
FJO
asked in iQ Suite
StatsIQ - Different correlation results if I use text set or number set

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

463
AmaraW
Former XM Community Team Member
2 years ago
J
jrdrinnon
asked in iQ Suite
Crosstabs and Randomization

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?

111
AmaraW
Former XM Community Team Member
2 years ago
A
QPN Level 1 ●
Apostolos99QPN Level 1 ●
asked in iQ Suite
Automate exports from crosstab?

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

111
AmaraW
Former XM Community Team Member
2 years ago
M
MoragB
asked in iQ Suite
How do i group a Likert question and use it in a report

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 various different users so th

7513
CarolK
Level 4 ●●●●
2 years ago
T
tiurino
posted in iQ Suite
Alternatives to MaxDiff on XM Platform

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 randomized), and the respondent would pi

651
K
Level 3 ●●●
2 years ago
T
TKaal
asked in iQ Suite
Conjoint Analyses complete question data

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

173
E
2 years ago
jp_blinkCX
jp_blinkCX
posted in iQ Suite
What algorithm does StatsIQ uses for Recommended Test?

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.

60
jp_blinkCX
2 years ago
algroom_heart
algroom_heart
asked in iQ Suite
Filtering and changing sentiment analysis in Text IQ

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?

541
Priyanka_Fid
Level 2 ●●
2 years ago
A
Abeer
asked in iQ Suite
Related questions and conditions

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?

41
jmborzick
QPN Level 5 ●●●●●
3 years ago
P
Phigam01
posted in iQ Suite
Anchored Max-Diff Methodology

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 items are actually dee

6161
CamM
Level 3 ●●●
3 years ago
C
Cindy
posted in iQ Suite
Conjoint: Responses not being counted

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?

80
C
3 years ago
Pradyp
Pradyp
asked in iQ Suite
No data available to analyze

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

3787
NataliaAssis
3 years ago
B
Ben99999
asked in iQ Suite
Standard exporter does not work for our survey but legacy exporter does

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 those write-in an

938
B
3 years ago
J
JYurkovich
asked in iQ Suite
Should near operators in TextiQ topic queries be symmetrical?

So a query that uses a near operator, for example:(celebrate difference)~2will return phrases like "celebrates our differences", "celebrate many differences", and "celebrating the differences", but not "differences are celebrated".However, this query will return "differences are celebrated"(difference celebrate)~2As will this query:(celebrate difference)~3So it appears that there is either a bug in how the distance is quantified, or a lack of documentation on the functionality. Either way, it doesn't make a lot of sense that it wouldn't be symmetrical.

281
J
3 years ago
H
hadard
asked in iQ Suite
Category type variables as input in regression

Hi,I'm working on finding key drivers using the stats IQ regression model. I wondered how the software calculates categorical parameters (e.g., Department) as predictors for numerical output (e.g., Overall CSAT). As far as I know, predictive parameters (inputs) in linear models must be either interval/ratio measurement scales or dummy variables.Does Stats IQ automatically code categorical variables into dummy variables in the regression model? Thanks for the help!

260
H
3 years ago
J
Jeff_J88
posted in iQ Suite
How to use conjoint simulator with only using imported data and no actual responses?

I recently ran a conjoint and after pulling the data into a report I realized we would need to add an attribute level to the model. I proceeded to add that level and re-field which caused the data I previously collected to not show up in the simulator. So I downloaded the raw data file, copied the survey and reverted the changes, then imported the data hoping the reports and simulator would work, but it is still telling me I have no responses. Anyone know how I can get my simulator to populate the data I already have without a complete re-field?Thanks in advance.

330
J
3 years ago
S
Sarka
asked in iQ Suite
Edited responses are not showing in Stat IQ

Hello, I had to edit a couple of responses but do not see them when running a Stat iQ analysis. What do I need to do to be able to see them? I am using a filter (for one category) and doing a simple "describe".Many thanks,Sarka

2163
Phillsh
Level 1 ●
3 years ago
T
TKaal
asked in iQ Suite
Conjoint analysis: what choice set did the respondent get to see?

Hi! I am working on estimating an MNL model and in order to do so, I need to know which choice set a respondent got to see. Does anyone know whether this is possible?

241
AmaraW
Former XM Community Team Member
3 years ago
AdamK12
Level 5 ●●●●●
AdamK12Level 5 ●●●●●
posted in iQ Suite
Any thoughts on multicollinearity among CX measures

Hello everyone! We just finished our regular check-in with Jake_Stack , and were wondering if anyone in the community has done any research or encountered any findings on multicollinearity between CX measures like net promoter score, overall satisfaction, task accomplishment, et cetera.It would be intuitive that these measures would be related, but given that we don't have a huge volume of data yet, I would be interested in exploring these relationships and seeing what other community members have been able to uncover. Thanks in advance!

131
AdamK12
Level 5 ●●●●●
3 years ago

Motivational Playlist

Terms & Conditions