Just filled out a web survey with a bunch of Likert-type items. The response categories were in the same order on each item, but not the order you'd expect:
fairly important
important
unimportant
very important
Pretty sure "alphabetize response categories" is not best practice in survey design.
fairly important
important
unimportant
very important
Pretty sure "alphabetize response categories" is not best practice in survey design.
October 6, 2025 at 6:42 PM
Just filled out a web survey with a bunch of Likert-type items. The response categories were in the same order on each item, but not the order you'd expect:
fairly important
important
unimportant
very important
Pretty sure "alphabetize response categories" is not best practice in survey design.
fairly important
important
unimportant
very important
Pretty sure "alphabetize response categories" is not best practice in survey design.
En ce qui concerne DORA, l'étude ne porte pas sur des entreprises qui ont réussie, mais sur des enquêtes Likert type. Y'a des assertions et les répondants disent s'ils sont plus ou moins d'acc avec. A partir de là les données sont compilées et des modèles d'équations structurelles ressortent
October 4, 2023 at 2:52 PM
En ce qui concerne DORA, l'étude ne porte pas sur des entreprises qui ont réussie, mais sur des enquêtes Likert type. Y'a des assertions et les répondants disent s'ils sont plus ou moins d'acc avec. A partir de là les données sont compilées et des modèles d'équations structurelles ressortent
#statstab #155 Power & Sample Size calculations for ordinal data {Hmisc}
Thoughts: An appropriate sample size is just as important as an appropriate (ordinal) analysis. Nice guide for a 2-group design with a Likert-type DV.
#ordinal #NHST #poweranalysis #clm
library.virginia.edu/data/article...
Thoughts: An appropriate sample size is just as important as an appropriate (ordinal) analysis. Nice guide for a 2-group design with a Likert-type DV.
#ordinal #NHST #poweranalysis #clm
library.virginia.edu/data/article...
August 9, 2024 at 6:06 PM
#statstab #155 Power & Sample Size calculations for ordinal data {Hmisc}
Thoughts: An appropriate sample size is just as important as an appropriate (ordinal) analysis. Nice guide for a 2-group design with a Likert-type DV.
#ordinal #NHST #poweranalysis #clm
library.virginia.edu/data/article...
Thoughts: An appropriate sample size is just as important as an appropriate (ordinal) analysis. Nice guide for a 2-group design with a Likert-type DV.
#ordinal #NHST #poweranalysis #clm
library.virginia.edu/data/article...
It's funny, most people would consider surveys with likert scale type items providing ordinal data as "quantitative"... Yet, Stevens held that only interval or ratio data are quantitative. So are likert scales really qualitative?
November 18, 2024 at 1:54 PM
It's funny, most people would consider surveys with likert scale type items providing ordinal data as "quantitative"... Yet, Stevens held that only interval or ratio data are quantitative. So are likert scales really qualitative?
Those bfi data already had missing values, and the focal variables were Likert-type, which allowed me to practice setting up an imputation strategy with the {mice} package. I then followed up the imputation with practicing the primary analyses, and how we'd present the results. It was very helpful.
April 15, 2025 at 3:05 PM
Those bfi data already had missing values, and the focal variables were Likert-type, which allowed me to practice setting up an imputation strategy with the {mice} package. I then followed up the imputation with practicing the primary analyses, and how we'd present the results. It was very helpful.
#SmartPeopleOfTwitter, what is this type of chart called? It is a really cool way to display Likert scale data.
December 1, 2024 at 8:27 AM
#SmartPeopleOfTwitter, what is this type of chart called? It is a really cool way to display Likert scale data.
#rstats #package #ShortForm is efficient to create #brief #shortform version of a #scale / #measurement that uses #Likert-type #anchors but there are alternatives rdrr.io/github/Antho..., here is the article journals.sagepub.com/doi/full/10....
October 5, 2025 at 6:33 AM
#rstats #package #ShortForm is efficient to create #brief #shortform version of a #scale / #measurement that uses #Likert-type #anchors but there are alternatives rdrr.io/github/Antho..., here is the article journals.sagepub.com/doi/full/10....
How to Work with Likert-Type Questionnaires? doi.org/10.1007/978-...
June 3, 2024 at 12:19 PM
How to Work with Likert-Type Questionnaires? doi.org/10.1007/978-...
It was a five point Likert-type scale response
June 9, 2024 at 6:53 PM
It was a five point Likert-type scale response
Likert-type data, due to their ordinal nature, can present challenges in the use of the median as a measure of central tendency. Since these data do not include decimal values, the median often falls within a limited range—typically 0 or 0.5.
April 12, 2025 at 12:30 PM
Likert-type data, due to their ordinal nature, can present challenges in the use of the median as a measure of central tendency. Since these data do not include decimal values, the median often falls within a limited range—typically 0 or 0.5.
We've heard (also anecdotally) that our own algorithm works well relative to manual flagging of participants, but it's only a matter of time before LLMs do the Likert-type items as well... so, I'm interested in hearing if it sounds like it's here already
June 25, 2025 at 3:10 AM
We've heard (also anecdotally) that our own algorithm works well relative to manual flagging of participants, but it's only a matter of time before LLMs do the Likert-type items as well... so, I'm interested in hearing if it sounds like it's here already
Not the biggest of dataset but nice,
In a dataset of 13 participants interviewed individually after 3 classic hypnosis scale suggestions describing each experience and rating their response in a Likert-type rating of 0-5, a LLM could zero-shot predict their rating from the transcript
r2=0.69
In a dataset of 13 participants interviewed individually after 3 classic hypnosis scale suggestions describing each experience and rating their response in a Likert-type rating of 0-5, a LLM could zero-shot predict their rating from the transcript
r2=0.69
June 10, 2025 at 12:50 PM
Not the biggest of dataset but nice,
In a dataset of 13 participants interviewed individually after 3 classic hypnosis scale suggestions describing each experience and rating their response in a Likert-type rating of 0-5, a LLM could zero-shot predict their rating from the transcript
r2=0.69
In a dataset of 13 participants interviewed individually after 3 classic hypnosis scale suggestions describing each experience and rating their response in a Likert-type rating of 0-5, a LLM could zero-shot predict their rating from the transcript
r2=0.69
The most common approach is an anchor-based approach (i.e., using an external anchor/reference variable as a guidance to derive a SESOI), especially for psychological variables that have no natural metric and are measured with Likert-type scales.
doi.org/10.1016/j.je...
Redirecting
doi.org
October 31, 2025 at 5:19 AM
The most common approach is an anchor-based approach (i.e., using an external anchor/reference variable as a guidance to derive a SESOI), especially for psychological variables that have no natural metric and are measured with Likert-type scales.
doi.org/10.1016/j.je...
arXiv:2503.04114v1 Announce Type: new
Abstract: Quadratic Surveys (QSs) elicit more accurate preferences than traditional methods like Likert-scale surveys. However, the cognitive load associated with QSs has hindered their adoption in [1/6 of https://arxiv.org/abs/2503.04114v1]
Abstract: Quadratic Surveys (QSs) elicit more accurate preferences than traditional methods like Likert-scale surveys. However, the cognitive load associated with QSs has hindered their adoption in [1/6 of https://arxiv.org/abs/2503.04114v1]
March 7, 2025 at 6:19 AM
arXiv:2503.04114v1 Announce Type: new
Abstract: Quadratic Surveys (QSs) elicit more accurate preferences than traditional methods like Likert-scale surveys. However, the cognitive load associated with QSs has hindered their adoption in [1/6 of https://arxiv.org/abs/2503.04114v1]
Abstract: Quadratic Surveys (QSs) elicit more accurate preferences than traditional methods like Likert-scale surveys. However, the cognitive load associated with QSs has hindered their adoption in [1/6 of https://arxiv.org/abs/2503.04114v1]
#statstab #404 {latent2likert} simulate Likert response variables from hypothetical latent variables
Thoughts: Most of psych is Likert type data. This R pkg can help simulate effects and check model fit.
#likert #ordinal #r #latent #simulation #data
latent2likert.lalovic.io/articles/usi...
Thoughts: Most of psych is Likert type data. This R pkg can help simulate effects and check model fit.
#likert #ordinal #r #latent #simulation #data
latent2likert.lalovic.io/articles/usi...
Using latent2likert
latent2likert
latent2likert.lalovic.io
August 21, 2025 at 7:36 PM
#statstab #404 {latent2likert} simulate Likert response variables from hypothetical latent variables
Thoughts: Most of psych is Likert type data. This R pkg can help simulate effects and check model fit.
#likert #ordinal #r #latent #simulation #data
latent2likert.lalovic.io/articles/usi...
Thoughts: Most of psych is Likert type data. This R pkg can help simulate effects and check model fit.
#likert #ordinal #r #latent #simulation #data
latent2likert.lalovic.io/articles/usi...
I think its a confluence of the skewness (which in my sim also messes with the tests in the presence of non-linearity) and the very abrupt non monotone conditional effect? Like if you sample normal shaped symmetric likert before, does the abrupt ifelse(before>=5) still mess up the type-1 error?
July 3, 2024 at 10:18 AM
I think its a confluence of the skewness (which in my sim also messes with the tests in the presence of non-linearity) and the very abrupt non monotone conditional effect? Like if you sample normal shaped symmetric likert before, does the abrupt ifelse(before>=5) still mess up the type-1 error?
If you do survey research with Likert-type items and worry about random responders, Michael John Ilagan & I made a Shiny app we think some of you may like.
falkcarl.shinyapps.io/BotApp1/
#SurveyResearch #DataQuality #Psychometrics #OnlineSurveys #SurveyBots #RandomResponders #rstats #Shiny
falkcarl.shinyapps.io/BotApp1/
#SurveyResearch #DataQuality #Psychometrics #OnlineSurveys #SurveyBots #RandomResponders #rstats #Shiny
July 28, 2025 at 2:13 PM
If you do survey research with Likert-type items and worry about random responders, Michael John Ilagan & I made a Shiny app we think some of you may like.
falkcarl.shinyapps.io/BotApp1/
#SurveyResearch #DataQuality #Psychometrics #OnlineSurveys #SurveyBots #RandomResponders #rstats #Shiny
falkcarl.shinyapps.io/BotApp1/
#SurveyResearch #DataQuality #Psychometrics #OnlineSurveys #SurveyBots #RandomResponders #rstats #Shiny
am building a survey in MS Teams and while writing the question for a potential Likert scale type set of questions, the AI suggestion just flat out suggested "Don't know" above the options 😂
Clicked it to see what would happen, and Literally Nothing Happened (except the button went away) lmao
Clicked it to see what would happen, and Literally Nothing Happened (except the button went away) lmao
October 30, 2025 at 4:14 AM
am building a survey in MS Teams and while writing the question for a potential Likert scale type set of questions, the AI suggestion just flat out suggested "Don't know" above the options 😂
Clicked it to see what would happen, and Literally Nothing Happened (except the button went away) lmao
Clicked it to see what would happen, and Literally Nothing Happened (except the button went away) lmao
New 5-point Likert-type item just dropped.
The exclamation points are sending me.
December 6, 2023 at 4:50 PM
New 5-point Likert-type item just dropped.
Type of therapist that wants to know stuff on a scale of 1-10, whether I Likert or not.
April 21, 2025 at 9:16 PM
Type of therapist that wants to know stuff on a scale of 1-10, whether I Likert or not.
Yes well I took their little “how did we do today” form and added a new column to the Likert-type scale of excellent-poor, labelled “atrocious” (my dad helped with the spelling), and shared my thoughts about our lost luggage and poor choice of snacks on that flight
October 12, 2023 at 10:46 AM
Yes well I took their little “how did we do today” form and added a new column to the Likert-type scale of excellent-poor, labelled “atrocious” (my dad helped with the spelling), and shared my thoughts about our lost luggage and poor choice of snacks on that flight
I'm using the Consultation and Relational Empathy [CARE] measure, a Likert-type scale that looks at compassion from the patient perspective. On the treatment side, CARE scores have been associated with better diabetes self-management outcomes & even all-cause mortality.
October 16, 2023 at 7:02 PM
I'm using the Consultation and Relational Empathy [CARE] measure, a Likert-type scale that looks at compassion from the patient perspective. On the treatment side, CARE scores have been associated with better diabetes self-management outcomes & even all-cause mortality.
I use "What do you think of Donald Trump" as an example of how to get an SD close to 3 from a 1–7 Likert-type item.
Headed right back toward the low 40s where he spent most of his first term. Not some electoral juggernaut. Not even popular.
At this point in Joe Biden's first term, his approval/disapproval was 53.2%/38.4% (+14.7). Trump is at 47.6%/47.9% (-0.3).
At this point in Joe Biden's first term, his approval/disapproval was 53.2%/38.4% (+14.7). Trump is at 47.6%/47.9% (-0.3).
March 5, 2025 at 2:33 PM
I use "What do you think of Donald Trump" as an example of how to get an SD close to 3 from a 1–7 Likert-type item.
As to the domain knowledge, I’ll use some recent use cases on projects I work on. My DVs are often Likert-type data, both items and sum scores. They often have missing data issues (especially clinical data).
April 15, 2025 at 3:05 PM
As to the domain knowledge, I’ll use some recent use cases on projects I work on. My DVs are often Likert-type data, both items and sum scores. They often have missing data issues (especially clinical data).
Some people might even point out that a Likert scale is a specific type of multi item scale and not any rating scale. Some people…
October 12, 2023 at 5:20 PM
Some people might even point out that a Likert scale is a specific type of multi item scale and not any rating scale. Some people…