Your favorite rating interface

13 Dec 2007 - 6:54pm
6 years ago
20 replies
5575 reads
Kim McGalliard
2006

Don't get me wrong, I love stars, but I feel like the 5 star rating
interface is getting a little old. Especially because these kind of ratings
average out to be in the middle over time and therefore aren't very
informative.

Anyone seen any interesting rating interfaces lately? I'm especially
interested in rating of CONTENT on a site (more than an ITEM you might buy
or have bought).

Thanks for any links. I can compile and send out to the list if people seem
interested.

Kim

Comments

13 Dec 2007 - 7:58pm
Robert Hoekman, Jr.
2005

> Anyone seen any interesting rating interfaces lately? I'm especially
> interested in rating of CONTENT on a site (more than an ITEM you might buy
> or have bought).

The one currently in use on Amazon product pages is actually pretty nice. It
breaks down the ratings with bar charts so you can se how many people rated
something a 5, 4, etc. It adds more meaning to the rating.

-r-

14 Dec 2007 - 2:03am
Jeffrey D. Gimzek
2007

Kim Asked:
>> Anyone seen any interesting rating interfaces lately? I'm especially
>> interested in rating of CONTENT on a site (more than an ITEM you
>> might buy
>> or have bought).

> The one currently in use on Amazon product pages is actually pretty
> nice. It
> breaks down the ratings with bar charts so you can se how many
> people rated
> something a 5, 4, etc. It adds more meaning to the rating.

I am working on a rating site now and let me tell you this is a tough
nut to crack !

1 - 5 stars are ubiquitous for a reason.

One scale i really liked was a branded one:

this site <http://www.mtbr.com/reviews/Saddle/product_22653.shtml>
used chili peppers.
I have seen less... socially acceptable scale icons in certain
magazines run by Hugh Hefner.

Ebay has a nice rating interface on their buyer/seller feedback that
allows them to use different Likert scales <http://en.wikipedia.org/
wiki/Likert_scale> with the same graphic 1 - 5 star graphic. ie: Like
- Dislike on one scale, Satisified to Unsatisfied on another.

the quality and quantity of the data is important - if you rate 5
things from 1-5, is the overall rating a 4 or a 4.5 or a 4.2 or even
4.25 ?

how important is that accuracy to your users?

Another aspect of rating is "helpful/unhelpful" where 27 helpful vs.
3 unhelpful basically "rate" a post... thumbs up/dn is also used this
way.

Sites such as Yelp and thefunded.com use this as an addition to rating.

We wound with a sort of LCD/LED meter like you see on equalizers, but
horizontal.
that way we can change the scale:

1 - 5
1.0 - 5.0
1 - 10
0 -100%

without really changing the graphic... what that meant however is
that we needed a real number next that gives you actual data: 3, 2.5,
7 of 10 etc.

sort of like this: || || || || | (4.5)

Again, this is a great place to brand your rating to make it unique.

hope that all helps you get started !

jd

--

Jeff Gimzek | Senior User Experience Designer

jeff at glassdoor.com | www.glassdoor.com

14 Dec 2007 - 6:42am
bminihan
2007

Thanks a ton for these, Jeffrey. I'm building ratings into our site, as
well, and one of my challenges is coming up with different ratings for
different kinds of people. We have three sets of users on our site -
coaches, athletes and fans, and I want to differentiate ratings from coaches
and fans (because one may have more "weight" than others in helping an
athlete get recruited), so I might use some of your ideas to do this.

Another rating interface I like (tho I've only looked at it a little bit) is
RottenTomatoes.com.

Thanka =]

Bryan
http://www.bryanminihan.com

-----Original Message-----
From: discuss-bounces at lists.interactiondesigners.com
[mailto:discuss-bounces at lists.interactiondesigners.com] On Behalf Of Jeffrey
D. Gimzek
Sent: Friday, December 14, 2007 2:03 AM
To: IXDA list
Subject: Re: [IxDA Discuss] Your favorite rating interface

Kim Asked:
>> Anyone seen any interesting rating interfaces lately? I'm especially
>> interested in rating of CONTENT on a site (more than an ITEM you
>> might buy
>> or have bought).

> The one currently in use on Amazon product pages is actually pretty
> nice. It
> breaks down the ratings with bar charts so you can se how many
> people rated
> something a 5, 4, etc. It adds more meaning to the rating.

I am working on a rating site now and let me tell you this is a tough
nut to crack !

1 - 5 stars are ubiquitous for a reason.

One scale i really liked was a branded one:

this site <http://www.mtbr.com/reviews/Saddle/product_22653.shtml>
used chili peppers.
I have seen less... socially acceptable scale icons in certain
magazines run by Hugh Hefner.

Ebay has a nice rating interface on their buyer/seller feedback that
allows them to use different Likert scales <http://en.wikipedia.org/
wiki/Likert_scale> with the same graphic 1 - 5 star graphic. ie: Like
- Dislike on one scale, Satisified to Unsatisfied on another.

the quality and quantity of the data is important - if you rate 5
things from 1-5, is the overall rating a 4 or a 4.5 or a 4.2 or even
4.25 ?

how important is that accuracy to your users?

Another aspect of rating is "helpful/unhelpful" where 27 helpful vs.
3 unhelpful basically "rate" a post... thumbs up/dn is also used this
way.

Sites such as Yelp and thefunded.com use this as an addition to rating.

We wound with a sort of LCD/LED meter like you see on equalizers, but
horizontal.
that way we can change the scale:

1 - 5
1.0 - 5.0
1 - 10
0 -100%

without really changing the graphic... what that meant however is
that we needed a real number next that gives you actual data: 3, 2.5,
7 of 10 etc.

sort of like this: || || || || | (4.5)

Again, this is a great place to brand your rating to make it unique.

hope that all helps you get started !

jd

--

Jeff Gimzek | Senior User Experience Designer

jeff at glassdoor.com | www.glassdoor.com

________________________________________________________________
*Come to IxDA Interaction08 | Savannah*
February 8-10, 2008 in Savannah, GA, USA
Register today: http://interaction08.ixda.org/

________________________________________________________________
Welcome to the Interaction Design Association (IxDA)!
To post to this list ....... discuss at ixda.org
Unsubscribe ................ http://www.ixda.org/unsubscribe
List Guidelines ............ http://www.ixda.org/guidelines
List Help .................. http://www.ixda.org/help

14 Dec 2007 - 8:19am
Todd Warfel
2003

> Kim Asked:
>>> Anyone seen any interesting rating interfaces lately? I'm
>>> especially interested in rating of CONTENT on a site (more than an
>>> ITEM you might buy or have bought).

Check out the one at Fancast http://www.fancast.com/movies/I-Am-Legend/95402/main
. It combines stars with regular language like: dreadful, awful, so
so, eh, pretty good and amazing.

Cheers!

Todd Zaki Warfel
President, Design Researcher
Messagefirst | Designing Information. Beautifully.
----------------------------------
Contact Info
Voice: (215) 825-7423
Email: todd at messagefirst.com
AIM: twarfel at mac.com
Blog: http://toddwarfel.com
----------------------------------
In theory, theory and practice are the same.
In practice, they are not.

14 Dec 2007 - 9:49am
Mihira Jayasekera9
2007

Bret Victor, in his treatise on interaction design, mentions a very simple way of improving the display of 5-star ratings by including "whiskers" under each star to indicate proportionally how ratings were distributed. (This is like a sparkline-ified version of Amazon's rating graph that Robert mentioned.)

http://worrydream.com/MagicInk/#p115

-Mihira

PS - Not to get off topic, but, incidentally, the whole paper is very provocative and well worth the read. Victor says in the intro: "I argue that interactivity is actually a curse for users and a crutch for designers, and users%u2019 goals can be better satisfied through other means."

14 Dec 2007 - 4:16pm
cfmdesigns
2004

On Dec 13, 2007, at 11:03 PM, Jeffrey D. Gimzek
(listserv at jdgimzek.com) wrote:

> I am working on a rating site now and let me tell you this is a tough
> nut to crack !
>
> 1 - 5 stars are ubiquitous for a reason.

The only reason 1-5 stars is ubiquitous is because it's ubiquitous.

Anyone remember when hotels and such only went to 4 stars? And now
I've seen ones touted as "7 stars". Sounds like "starflation" to me:
when "everyone" is a 4-star, you have to go to a 5-star system just to
differentiate yourself from the crowd. But why is everyone a 4-star?
Answer: not because they are all top of the line, but because no one
wants to be below the top. (And doubly so when Motel 6 and its peers
show up as 2-star, leaving the question of what fleabag is only a 1-
star? Eew?

The problem of a 1-5 star system is that there is typically no
definition of what the various ratings mean. Is 3-star "average"?
(What is "average"?) Is 1-star "avoid even if they pay you to stay
there and it's the middle of a blizzard and there is no other place at
all within 100 miles"? Does no stars mean "hasn't been rated" or
"less than 1 star"; does it mean both?

A number of systems have gone to a 7-star system: 5 stars, no stars,
and "don't like". (Adobe Bridge, Rhapsody, many others.) Even then,
it may do a good job of capturing levels of dislike which may be
valuable in some settings -- movies, for example. For that, an 8-star
system is probably better: -3 -2 -1 0 1 2 3 (and unrated). Then you
could decide that (picking two movies I saw from the Rotten Tomatoes
"worst" list from last year) than while DaVinci Code and Eragon were
both "bad", that Eragon was a -2 (Tivo it and maybe remember to watch
it later, or not) but DaVinci was only a -1 (might be worth Netflixing
some day), while Dreamgirls was a +2 (might be worth buying a copy for
your personal library). (Such a positive/negative balance also makes
rolling up group averages more accurate.)

-- Jim Drew
cfmdesigns at earthlink.net
http://www.soundskinky.com/blog/

14 Dec 2007 - 5:00pm
Jeffrey D. Gimzek
2007

On Dec 14, 2007, at 1:16 PM, Jim Drew wrote:

>
> On Dec 13, 2007, at 11:03 PM, Jeffrey D. Gimzek
> (listserv at jdgimzek.com) wrote:
>
>> I am working on a rating site now and let me tell you this is a tough
>> nut to crack !
>>
>> 1 - 5 stars are ubiquitous for a reason.
>
> The only reason 1-5 stars is ubiquitous is because it's ubiquitous.
>
> Anyone remember when hotels and such only went to 4 stars? And now
> I've seen ones touted as "7 stars". Sounds like "starflation" to me:
> when "everyone" is a 4-star, you have to go to a 5-star system just to
> differentiate yourself from the crowd. But why is everyone a 4-star?
> Answer: not because they are all top of the line, but because no one
> wants to be below the top. (And doubly so when Motel 6 and its peers
> show up as 2-star, leaving the question of what fleabag is only a 1-
> star? Eew?
>
>
> The problem of a 1-5 star system is that there is typically no
> definition of what the various ratings mean. Is 3-star "average"?
> (What is "average"?) Is 1-star "avoid even if they pay you to stay
> there and it's the middle of a blizzard and there is no other place at
> all within 100 miles"? Does no stars mean "hasn't been rated" or
> "less than 1 star"; does it mean both?
>
> A number of systems have gone to a 7-star system: 5 stars, no stars,
> and "don't like". (Adobe Bridge, Rhapsody, ma ny others.) Even then,
> it may do a good job of capturing levels of dislike which may be
> valuable in some settings -- movies, for example. For that, an 8-star
> system is probably better: -3 -2 -1 0 1 2 3 (and unrated). Then you
> could decide that (picking two movies I saw from the Rotten Tomatoes
> "worst" list from last year) than while DaVinci Code and Eragon were
> both "bad", that Eragon was a -2 (Tivo it and maybe remember to watch
> it later, or not) but DaVinci was only a -1 (might be worth Netflixing
> some day), while Dreamgirls was a +2 (might be worth buying a copy for
> your personal library). (Such a positive/negative balance also makes
> rolling up group averages more accurate.)

OK, but is that instantly understandable ?

that is the deal with the stars - ubiquitous = understandable

many many people dont really understand negative numbers

to see an example of a rating that is almost totally useless, see
jobvent.com

what's a 1303? what's a - 43? i dont know. i have to figure it out,
and by then i dont care.

jd

- -

Jeffrey D. Gimzek | Senior User Experience Designer

http://www.glassdoor.com

14 Dec 2007 - 4:46pm
kimbieler
2007

Clearly, the Michelin guide's ratings system is superior, since they
only have 3 stars. <g> Of course, 99.99% of restaurants are beneath
their notice and receive no star, so they're only providing ratings
from excellent to superlative. Which is actually a good way of
controlling your ratings.

If I'm being asked to give ratings, I tend to prefer a system that's
spelled out. For example:

1 -- I'll never shop here again, you suck
2 -- It would take a major sale to convince me to shop here again
3 -- I will probably shop here again
4 -- I'll definitely shop here again, and recommend you to friends
5 -- Please, take all my money, I love you so much

I think spelling out the ratings prevents me from having to think too
hard about it ("hmm, is it a 6 or a 7?"). You might also shame people
into not giving you such a lousy rating by using over-the-top humor.

Movie/book ratings are always hard because you can either rate all
movies against each other or you can rate within genre. So, Eragon
might get a 1/10 compared to movies overall, but a 4/10 compared to
other fantasy movies. I've also noticed that I tend to rate movies
higher if I had low expectations and lower if I had high
expectations. One might get better results if you asked people to
rate movies compared to other movies. That way you can control
whether they're within genre, for example.

-- Kim

+ + + + + + + + + + + + + + + + + +
Kim Bieler Graphic Design
www.kbgd.com
240-476-3129
+ + + + + + + + + + + + + + + + + +

15 Dec 2007 - 11:36pm
Jonathan Koren
2007

Jim hit on the head. Stars are ubiquitous because they're
ubiquitous. People expect them, and they're deceptively simple.

It's hard to capture what a rating actually means. Take a simple
thing like, "Rate this movie from 1 (lowest) to 5 (highest)." What
does that mean? Does it mean "Rate the quality of this movie?" or
does it mean "Rate how much you enjoyed this movie?" There's plenty
of movies that are obviously well made, well acted, and of undeniably
high quality, that I simply just don't like. Similarly, there's
plenty of crap movies that I enjoy immensely. How do I rank these?
Sure you could have two different scales, but as soon as you add a
second, you're quickly going to want to add a third and fourth, until
the rating system becomes overwhelming.

Even a Likert scale doesn't mean the same thing to users. There's
plenty of users that rank everything 5 stars, others that rank items
only 1 star or 5 stars. Even in my own experience a Likert scale can
be problematic. I decided to rate all my songs on iTunes. I decided
I'd try to bell curve it with:

1: absolute crap, or non music track
2: don't like it
3: perfectly acceptable
4: like it
5: sublime

This worked for a while, until I started to realized that there would
be two songs that I liked, but weren't sublime, but I preferred one to
the other. I couldn't capture that relationship. In a sense, a 7
point scale would have been better here, but I suspect I would have
ended up with wanting a 9 point scale.

Typically, people just use a 5 point Likert scale and tell users, "do
what you want with it" and then renormalize the rating across all the
users before doing something interesting with it.

Personally, my experience with Likert scales has made me appreciate
the simple "thumbs up"/"thumbs down" approach Tivo has taken. :)

--
Jonathan Koren
jonathan at soe.ucsc.edu
http://www.soe.ucsc.edu/~jonathan/

16 Dec 2007 - 1:59am
cfmdesigns
2004

On Dec 14, 2007, at 2:00 PM, Jeffrey D. Gimzek wrote:

>> A number of systems have gone to a 7-star system: […] Even then,
>> it may do a good job of capturing levels of dislike which may be
>> valuable in some settings -- movies, for example. For that, an 8-
>> star
>> system is probably better: -3 -2 -1 0 1 2 3 (and unrated). […]
>
>
> OK, but is that instantly understandable ?
>
> that is the deal with the stars - ubiquitous = understandable

Ubiquitous != understandable. Just because something is everywhere,
doesn't mean everyone understands it in the same way. As stated,
that's the exact problem with a 5-star system. Everyone probably
understands what 5 stars means, but all the rest are up for grabs.

> many many people dont really understand negative numbers
>
> to see an example of a rating that is almost totally useless, see
> jobvent.com
>
> what's a 1303? what's a - 43? i dont know. i have to figure it out,
> and by then i dont care.

I can't disagree with you; a (presumably) rolled-up value which
doesn't obviously relate to the original rating spread is obscure at
best, maybe unuseful. One which does relate -- average 3.4 rating (19
ratings) on a scale of -4 to +4 -- is better.

If negative numbers are deemed too confusing -- where're your
personas, huh?! <grin> -- maybe a letter grade of A through F with +
and - in the mix. Americans at least should have a decent grasp that
if the grade relates to percentages, then an F is covering roughly the
entire bottom half of the spectrum, everything from completely
unacceptable down to putrid and beyond.

-- Jim Drew
cfmdesigns at earthlink.net
http://www.soundskinky.com/blog/

16 Dec 2007 - 2:56am
Murli Nagasundaram
2007

Probably ubiquitous in North America, and perhaps in Western Europe.
Certainly, it isn't too hard for people from other cultures to become
socialized into using such a system. Likert and other such interval scales
have been around for at least two generations in North America and their use
has become near universal in that culture (down to McDonald's restaurant
1-minute surveys). The influence of science as well as scientism (as in
Christian Science, Scientology and Intelligent Design) is very strong in
North American culture and the hard sciences demand quantification of just
about any measure. Scales are introduced to children in pre-school -- and so
are stars.
I am not sure that my mother, who was raised in India, and despite her
exposure to Western culture (and even after having visited a variety of
countries) would be able to use a star rating scale automatically.

The other problem is the implicit criteria used in a star rating as both
Jonathan and Jim point out.

What's an alternative though, that is simple and quick, as well as reliable?

-m

On 12/16/07, Jonathan Koren <jonathan at soe.ucsc.edu> wrote:
>
>
> Jim hit on the head. Stars are ubiquitous because they're
> ubiquitous. People expect them, and they're deceptively simple.

murli nagasundaram, ph.d. | www.murli.com | murli at murli.com | +91 99 02 69
69 20

16 Dec 2007 - 10:54am
Jonathan Koren
2007

On Dec 15, 2007, at 11:56 PM, Murli Nagasundaram wrote:

> The other problem is the implicit criteria used in a star rating as
> both Jonathan and Jim point out.
>
> What's an alternative though, that is simple and quick, as well as
> reliable?

:-C :-( :-| :-) :-D

;-)

--
Jonathan Koren
jonathan at soe.ucsc.edu
http://www.soe.ucsc.edu/~jonathan/

16 Dec 2007 - 1:16pm
bminihan
2007

I have a problem with ratings in iTunes, as well. I've grown a couple
hundred songs in my library over time, but about 98% of my library is either
not rated or 5. I wouldn't really have any 1s in iTunes, or it wouldn't be
there, so that rules out one whole point in the scale.

What would be really nice is if they captured the relationship of different
songs to each other. I tend to listen to iTunes in "sessions", so when I
sit down for a few hours to work, I play "work songs". I have a work
playlist, but it's incomplete, and I just don't have time to keep moving new
songs into it. I also want different types of music depending on what kind
of work I'm doing, and can't think of obvious names for different lists. If
iTunes could somehow pay attention to what songs I play multiple times, and
which songs I skip, in a given session, then let me play that as a playlist,
I wouldn't have to manually rate them or move them over to a list myself.
Over time iTunes would have compilations based on different moods and could
grow them by recognizing when I'm playing songs out of a particular list.

Wishful thinking, I know =]

Bryan
http://www.bryanminihan.com

-----Original Message-----
From: discuss-bounces at lists.interactiondesigners.com
[mailto:discuss-bounces at lists.interactiondesigners.com] On Behalf Of
Jonathan Koren
Sent: Saturday, December 15, 2007 11:37 PM
Cc: ixd-discussion
Subject: Re: [IxDA Discuss] Your favorite rating interface

Even a Likert scale doesn't mean the same thing to users. There's
plenty of users that rank everything 5 stars, others that rank items
only 1 star or 5 stars. Even in my own experience a Likert scale can
be problematic. I decided to rate all my songs on iTunes. I decided
I'd try to bell curve it with:

1: absolute crap, or non music track
2: don't like it
3: perfectly acceptable
4: like it
5: sublime

This worked for a while, until I started to realized that there would
be two songs that I liked, but weren't sublime, but I preferred one to
the other. I couldn't capture that relationship. In a sense, a 7
point scale would have been better here, but I suspect I would have
ended up with wanting a 9 point scale.

--
Jonathan Koren
jonathan at soe.ucsc.edu
http://www.soe.ucsc.edu/~jonathan/

16 Dec 2007 - 3:23pm
Michael Micheletti
2006

There's another scheme in work on Pandora (http://www.pandora.com). It seems
to work pretty well for tuning a "station" (music stream). There are two
basic rating icons/choices, plus three other related actions associated with
the rating system and station. The choices are identified by the (icon)
images listed below.

Ratings:
(Thumb Up) "I really like this song - play more like it!"
(Thumb Down) "I don't like this song - it's not what this station should
play."

Related:
(zzz) "I'm tired of this song - don't play it for a month."
(?) "Why is this song playing?"
(+) "I want to add more kinds of music to this station."

Pandora is one of the rare websites/web services that all the demographics
in our house love. I think that's partly because it's so easy to rate
content. Thumb's up or thumb's down.

Michael Micheletti

On Dec 16, 2007 10:16 AM, Bryan Minihan <bjminihan at nc.rr.com> wrote:

> What would be really nice is if they captured the relationship of
> different
> songs to each other. I tend to listen to iTunes in "sessions", so when I
> sit down for a few hours to work, I play "work songs".
>

17 Dec 2007 - 7:39pm
Oleh Kovalchuke
2006

Pandora has good rating system. However there is no way to change song
rating from "Thumbs Up" to "No Digit" rating. Since the rating affects the
frequency of the song being played, (for the upcoming year) I wish Santa has
added the neutral rating: "I don't mind this song, but I do not want it to
be played too frequently".

Oleh
On Dec 16, 2007 1:23 PM, Michael Micheletti <michael.micheletti at gmail.com>
wrote:

> There's another scheme in work on Pandora (http://www.pandora.com). It
> seems
> to work pretty well for tuning a "station" (music stream). There are two
> basic rating icons/choices, plus three other related actions associated
> with
> the rating system and station. The choices are identified by the (icon)
> images listed below.
>
> Ratings:
> (Thumb Up) "I really like this song - play more like it!"
> (Thumb Down) "I don't like this song - it's not what this station should
> play."
>
> Related:
> (zzz) "I'm tired of this song - don't play it for a month."
> (?) "Why is this song playing?"
> (+) "I want to add more kinds of music to this station."
>
> Pandora is one of the rare websites/web services that all the demographics
> in our house love. I think that's partly because it's so easy to rate
> content. Thumb's up or thumb's down.
>
> Michael Micheletti
>
> On Dec 16, 2007 10:16 AM, Bryan Minihan <bjminihan at nc.rr.com> wrote:
>
> > What would be really nice is if they captured the relationship of
> > different
> > songs to each other. I tend to listen to iTunes in "sessions", so when
> I
> > sit down for a few hours to work, I play "work songs".
> >
> ________________________________________________________________
> *Come to IxDA Interaction08 | Savannah*
> February 8-10, 2008 in Savannah, GA, USA
> Register today: http://interaction08.ixda.org/
>
> ________________________________________________________________
> Welcome to the Interaction Design Association (IxDA)!
> To post to this list ....... discuss at ixda.org
> Unsubscribe ................ http://www.ixda.org/unsubscribe
> List Guidelines ............ http://www.ixda.org/guidelines
> List Help .................. http://www.ixda.org/help
>

--
Oleh Kovalchuke
Interaction Design is the Design of Time
http://www.tangospring.com/IxDtopicWhatIsInteractionDesign.htm

18 Dec 2007 - 1:15pm
Jenni Merrifield
2007

Jeffrey D. Gimzek <listserv at jdgimzek.com> wrote:
> We wound with a sort of LCD/LED meter like you see on equalizers, but
> horizontal.
> that way we can change the scale:
>
> 1 - 5
> 1.0 - 5.0
> 1 - 10
> 0 -100%
>
> without really changing the graphic... what that meant however is
> that we needed a real number next that gives you actual data: 3, 2.5,
> 7 of 10 etc.
>
> sort of like this: || || || || | (4.5)

This reminds me of the rating method used at www.TV.com which I like.

If you are registered with the site, you can rate any show using an
"equalizer bar" that is initially coloured grey but as you move the rating
selector each bar lights up with colour, starting with red and progressing
to green the better you rate it. Although displayed as a 10-point scale,
it's actually a 100-point scale as you can choose values down to the closest
0.1. The overall rating is displayed as a decimal number that (I presume)
is the average rating (i.e. 8.3 or 5.2) but your personal rating is always
shown with the bar and can be modified at any time.

The aspect of the mechanism for selecting a personal rating that I like the
most is the colour coding of the bars for your selection. I'd also like to
see the same colour coding used for the overall rating - that is, show the
number in the appropriate colour, or display a thinner bar with the length
and red-to-green values for that number.

:-j(enni)

~~~~~
jenni merrifield -- jenniferm at cgtvgames.com
CGTV Games -- The Power of Interactive Entertainment
~~~~~
Designing to requirements
And walking on water
Are easy if both are frozen
~~~~~

18 Dec 2007 - 4:17pm
Esteban Barahona
2006

For reviewing content it's preferable to have reviews than ratings...
if the point is to improve it. And after using an account on
wikipedia, it's preferable being able to change directly the
content.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Posted from the new ixda.org
http://gamma.ixda.org/discuss?post=23493

19 Dec 2007 - 6:29am
prototypeangel
2007

I think the author just wants to find new ways for users to give quick
feedback. I personally like the rating system on ign and gamespot and
there are a couple of other interesting rating systems I quite can't
remember. One was an interesting wordpress plugin. I can't quite
remember the names of either

On Dec 19, 2:17 am, Esteban Barahona <esteban.barah... at gmail.com>
wrote:
> For reviewing content it's preferable to have reviews than ratings...
> if the point is to improve it. And after using an account on
> wikipedia, it's preferable being able to change directly the
> content.
>
> . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
> Posted from the new ixda.orghttp://gamma.ixda.org/discuss?post=23493
>
> ________________________________________________________________
> *Come to IxDA Interaction08 | Savannah*
> February 8-10, 2008 in Savannah, GA, USA
> Register today:http://interaction08.ixda.org/
>
> ________________________________________________________________
> Welcome to the Interaction Design Association (IxDA)!
> To post to this list ....... disc... at ixda.org
> Unsubscribe ................http://www.ixda.org/unsubscribe
> List Guidelines ............http://www.ixda.org/guidelines
> List Help ..................http://www.ixda.org/help

19 Dec 2007 - 2:01am
cone johnson
2007

I've just recently been researching visual metaphors for one of our
sites and noted that Petco uses paw prints; Burpee uses leaves; but
strangely, Figleaves.com did not use fig leaves.

I found most sites use stars - in some variation or other.

And I agree with Esteban that reviews are actually more effective
that the ratings. Perhaps the ratings can just be considered
"short-hand" to the reviews which (hopefully) provide the all
important and much needed details.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Posted from the new ixda.org
http://gamma.ixda.org/discuss?post=23493

19 Dec 2007 - 11:44am
Kim McGalliard
2006

Thanks everyone for excellent examples and good discussion. We're
definitely thinking about requiring reviews with ratings, but I'm a little
concerned that we won't get the quantity of ratings that we would get if we
just had ratings. Granted, the QUALITY of the ratings might be higher, so
it's a trade off of sorts.
I think we may go with a slider of some sort since it gives a bigger range
of numbers and is just as easy (if not easier) to use and understand as the
ubiquitous stars.

Thanks,

Kim

Syndicate content Get the feed