Thumbs Up/Down Style Ratings
From Social Patterns
A user wants to express a like/dislike (love/hate) type opinion about an object (person, place or thing) they are consuming / reading / experiencing.
The benefit to other users is that these ratings, when assessed in aggregate, can quickly give a sense for the community's opinion of a rated object. They may also be helpful for drawing quick qualitative comparisons between like items (this is better than that) but this is of secondary importance with this ratings-type.
- Use when you want to provide the ability to quickly grab a user’s opinion on an object.
- Use as an easy, fun way to begin engaging users in the community.
- Use when polarized opinions are more appropriate for the experience than degrees of opinion.
Do’s and Don’ts
DO consider thumb ratings when opinions about a rated asset will tend to be strongly polarized. For example, if you can state the question as simply as "Did you like this, or did you not?" then thumbs may be appropriate. If it seems more natural to state the question as "How much did you like this?" then Star-ratings are probably more appropriate.
DO consider using thumb ratings for developing personalized recommendations.
For example, Pandora uses declared music interests to create a playlist of songs that are similar in style and then uses the thumbs up rating on a song to add more songs like this to the music stream and a thumbs down to remove that song from the music stream.
- Place the thumb rating widget in close proximity to the asset being rated.
- Make sure the rating widget is secondary in prominence to the primary call to action, unless rating is the primary task. For example, in a shopping context, it is probably appropriate to keep 'Add to Cart' for an item as the predominant call-to-action, with 'Rate This' being less noticeable.
- Be consistent in the treatment of the imagery used for thumbs across the site or group of sites if applicable.
- Indicate to the user whether or not they have previously rated the item.
- If possible, refresh the widget inline as a user votes and clearly indicate what was selected.
- Allow the user to change their vote at any time.
- Present community consensus tallies as whole numbers rather than percentages for simplicity and ease of understanding.
- Consider highlighting items that have reach a certain level of positive votes over a long amount of time by a large amount of people.
- Be cautious about promoting the lowest rated or least favored items as the negativity can appear insensitive or rude.
Vote Counts - why not percentages?
Percentage comparisons between 2 or more rated items are problematic, due to response liquidity: when there are a low number of ratings (as there will be, initially, for any rated item) small differences of opinion can yield big perceived differences in rating.
For example, if 2 of 3 users give Movie A a thumbs-down, it appears that community consensus is 66% negative, which is technically accurate, but may not really represent an accurate picture of the tastes of the larger community. Really, 2 out of 3 votes just don’t count for much.
However, if 666 out of 1,000 users slam Movie B (expressed as percentage, also 66%) then that probably is a significant indicator of the larger community's opinion about that item.
When viewing Movie A alongside Movie B in a feature on 'Hot Movies Opening this Weekend', it's misleading to represent their approval ratios as percentages. Better to just give our users the numbers (how many positive, how many negative, and how many overrall) and let them figure it out.
Note that this is not the ideal situation, which is why we recommend that any context that relies on heavy data comparisons (eg. the ability to sort, filter or promote content based on ratings) may not be an appropriate use of Thumb-style ratings.
Also Note: that this 'Liquidity Problem' is not unique to Thumb-ratings. (It is, in fact, a well-known factor in economics)
Thumbs Up Only
Use a thumbs up type rating only when you don’t want negative ratings to appear insensitive or inappropriate, such as rating a person.
A thumbs-up – positive only variation may be contextually-appropriate for the assets that you want people to rate. If you strongly suspect that opinions about a type of asset will concentrate toward the positive pole of opinion (consistently) then offer them only that option.
A thumbs-up – positive only variation may be more culturally-appropriate to present. In some cultures it is deemed less-than-appropriate to express a strong negative opinion about something. (And, remember, Thumb-ratings are recommended for contexts in which opinions are strongly polarized.) For these locales, it may be preferable to provide positive-only opinions, and just let the absence of a rating connote a negative opinion.
The concept of Thumb-ratings can be problematic for some of locales.
First, the symbology of an extended and raised thumb is problematic: in some locales, it is considered an insult; in other locales, representations of any body-part (especially disembodied ones) are considered offensive; and, finally, some locales will just not 'get it.'
Secondly, the very notion of a binary, black-or-white, love-it or hate-it ratings system may not be a natural fit for some cultures. Many locales have stated a preference for 'shades of gray' in a polarized scale. (Note this is subtly different than Star-ratings, which imply "I like it exactly this much.")
Third, in some areas it is considered rude to criticize things openly. In this instance, consider the Thumbs Up only variation.
As Seen On
- From the work of Bryce Glass and the Yahoo! Social Platform team
- The Value of Downvoting...