{"id":2906,"date":"2016-06-04T03:39:16","date_gmt":"2016-06-04T07:39:16","guid":{"rendered":"http:\/\/www.sogolytics.com\/blog\/?p=2906"},"modified":"2025-07-16T13:55:40","modified_gmt":"2025-07-16T17:55:40","slug":"when-why-and-how-to-use-weighted-scores-in-survey-rating-questions","status":"publish","type":"post","link":"https:\/\/www.sogolytics.com\/blog\/when-why-and-how-to-use-weighted-scores-in-survey-rating-questions\/","title":{"rendered":"When, Why, and How to Use Weighted Scores in Survey Rating Questions"},"content":{"rendered":"

Even the most basic survey tool lets you measure responses using a ranking or rating scale. For instance, it\u2019s common to use satisfaction or importance scales (e.g. \u201cvery important\u201d to \u201cvery unimportant\u201d) or recommendation scales (\u201cHow likely are you to recommend us to a friend?\u201d). Sometimes, however, your data analysis can benefit from applying weighted scores to the answers during the design phase.<\/p>\n

<\/div>\n

\"rating<\/p>\n

<\/div>\n

Why bother with rating questions?<\/h2>\n

A typical survey includes some kind of qualitative assessment. That is, you ask the respondent to grade a product, experience, or feature on a scale you provide: How important is this? How satisfied were you? How likely would you be to buy this?<\/i> And that works great, in the large majority of situations.<\/p>\n

<\/div>\n

Generally, when we conduct a survey<\/a> we\u2019re looking for an overall perception: How happy are our customers, and what can we do to make them happier? It isn\u2019t always necessary to fuss with the details. Keep it simple!<\/p>\n

<\/div>\n

However, in some circumstances, you can make better decisions by weighing the answers.<\/p>\n

<\/div>\n

Perhaps the best explanation is an example used in The Increasing Problem With the Misinformed<\/a>, which (among other things), uses PolitiFact data that measures the truthfulness of politicians\u2019 statements. The author, Thomas Baekdal, initially graphs politicians\u2019 accuracy using an ordinary rating scale. But then he points out: \u201c…By ranking the data like this, we don’t take into account the severity of the lies a person makes. A person who made 10 small lies will be ranked the same as a person who made 10 big lies. Both are obviously bad, but we should really punish people in relation to the severity of their lies.\u201d<\/p>\n

<\/div>\n

Instead, he uses a system based on a logarithmic scale:<\/p>\n

<\/div>\n

We give \u201chalf-true\u201d a value of 1 and center it on the graph.<\/p>\n

<\/div>\n

We give \u201cMostly True\u201d a value of 2, and \u201cTrue\u201d a value of 5. The idea here is that we reward not just that something is true, but also that it provides us with the complete picture (or close to it).<\/p>\n

<\/div>\n

Similarly, we punish falsehoods. So, \u201cMostly False\u201d is given a value of -2, and \u201cFalse\u201d a value of -5.<\/p>\n

<\/div>\n

Finally, we have intentional falsehoods, the \u201cPants on Fire,\u201d which we punish by giving it a value of -10.<\/p>\n

<\/div>\n

My point is not to highlight anything about politicians or the media, which is what most interests Baekdal, but rather to show that sometimes one answer is more important than another, and your data analysis should reflect that.<\/p>\n

<\/div>\n

A few more typical business examples:<\/p>\n

<\/div>\n