Say bye-bye to the Likert scale!
A really interesting research paper on an experiment comparing responses between a five-point Likert scale and a 250-value “visual analog scale.”
Randomly assigned participants spent the same amount of time on each, but those using the visual scale made finer adjustments to their responses, resulting in more precise data. If you have ever felt you’ve had to compromise between choosing a “3” or a “4” on a Likert scale, you can understand why this research matters.
Even more interesting is the description in the paper of a similar device from the early 19th century: “a metal plate […] had 10 scales, each marked off in 100 parts, and labeled […]. A system of sliding markers was provided so that a […] judged position on each scale could be graphically displayed.”
The authors of this paper argue that use of such visual scales permit the ”detection of small differences and far more possibilities for data analyses” through fine-grained web-based data collection.
Frederik Funke and Ulf-Dietrich Reips (In press). Why Semantic Differentials in Web-Based Research Should be Made From Visual Analogue Scales and Not From 5-Point Scales. (pdf)
135 Notes/ Hide
-
gomeznude2012 liked this
-
selena2088gomez liked this
-
troythulu reblogged this from visualturn
-
shasou reblogged this from visualturn and added:
<Science> <タイトル> リッカート尺度にさよなら! ...<訳(一部略)>...
-
suryaray reblogged this from visualturn
-
danpeterson2 reblogged this from visualturn
-
sit-look-wonder reblogged this from visualturn
-
This was featured in #Science
-
visualturn posted this
![Say bye-bye to the Likert scale!
A really interesting research paper on an experiment comparing responses between a five-point Likert scale and a 250-value “visual analog scale.”
Randomly assigned participants spent the same amount of time on each, but those using the visual scale made finer adjustments to their responses, resulting in more precise data. If you have ever felt you’ve had to compromise between choosing a “3” or a “4” on a Likert scale, you can understand why this research matters.
Even more interesting is the description in the paper of a similar device from the early 19th century: “a metal plate […] had 10 scales, each marked off in 100 parts, and labeled […]. A system of sliding markers was provided so that a […] judged position on each scale could be graphically displayed.”
The authors of this paper argue that use of such visual scales permit the ”detection of small differences and far more possibilities for data analyses” through fine-grained web-based data collection.
Frederik Funke and Ulf-Dietrich Reips (In press). Why Semantic Differentials in Web-Based Research Should be Made From Visual Analogue Scales and Not From 5-Point Scales. (pdf)](http://24.media.tumblr.com/tumblr_ls0o2tVGBR1qzowgeo1_r1_1280.jpg)
