Apple Removes 'Rate on Delete' for Apps in iPhone 4

[johnnyhow]

As an app buyer, I want to let you app developers know that I put more weight on user feed back than on the star rating system because it's clear that a lot of people don't know whether 1 star is the top rating, or 5. So often I've seen comments that give 1 star but are in fact praising the app.

As for the 'rate to delete' option, I almost always clicked no thanks, unless I felt the quality was really shoddy (and believe me, there's a lot of trash in the app store).

 

[cteel2004]

You are skewing logic here...

If people delete the app as you say most likely don't like it, so they should be able to give it a low rating. Why is this skewing towards low ratings? The apps that are liked and not deleted, don't get bad ratings, but the ones that are deleted, get the bad ratings. If these apps were any good, they would not be deleted and they wouldn't be getting bad ratings? So, ultimately the good apps that are not deleted get the better grades, and those that do get deleted the worst.

I think this not only does not skew ratings, it rectifies them.

Moronic reporting by macrumors too, who take this biased view without any criticism. This does decrease overall ratings, but it doesn't skew them in any way. The good apps don't get deleted and don't get the bad rating, simple as that, the bad ones do, and deservedly so.

This is a bad move, because now, an app can be very unsatisfactory and it can be deleted by many iphone users, yet none of them will be prompted to rate it, and so give the negative feedback back to the community. Only if the user is really motivated, and the app is atrocious will they explicitly go to the app store and rate it...

Bad, Bad move, now all those third rate apps that we delete from our phones won't get a just low rating, and they will escape for some other poor schmuck to download unbeknownst to them.

You're missing an important point. The rating that a potential customer sees on an app is perceived as being representative of all the customers that have tried an app (both those that liked it, and those that didnt).

But that wasnt the case in the old system. People deleting an app had more opportunity to rate an app (in the form of the popup) than the people keeping an app. So as a result, ratings from people that are unsatisfied (for any reason) are more likely to give a rating than those that are satisfied.

So, breaking it down to the math of the rating system. You could have 1000 customers who have used your app. 10 people didn't like it, and so they deleted it. While the remaining 990 people liked it and did keep the app. This is a very good situation as a developer. 99% satisfaction rate. But those 10 people who didn't like the app were asked for their opinion, and they gave it. But the 990 people who liked it never were asked for their opinion, and so never gave it. As a result, you've got in the AppStore 10 votes with one star, and no votes with 3, 4, or 5 stars. So your average rating in the eyes of a potential new customers is 1 star, and the potential new customer will think 'hmm, doesn't look good. i'll move on to something else.'

It was more like the 'rate your service' cards at a restaraunt. You never fill those out unless you've had a bad experience.

Now, if Apple were to prompt customers who kept their copy of the app as much as they prompted customers who deleted their copy, then you would have a better picture of customer satisfaction with the app.

 

[Jeremy1026]

Ah, I will miss this great opportunity to rate down apps that are annoying, keep crashing and so on. But in the meantime, I'm on a jailbroken iPhone 3G. WHY WOULD I UPGRADE TO 4.0?!

If you need to rate an app poorly, then do so in iTunes. You don't need to be prompted to rate an app.

 

[Jeremy1026]

Moronic reporting by macrumors too, who take this biased view without any criticism. This does decrease overall ratings, but it doesn't skew them in any way. The good apps don't get deleted and don't get the bad rating, simple as that, the bad ones do, and deservedly so.

The good apps also don't get good ratings for being on the device. The user is never prompted to rate an app if it is left on the device. How is that not skewing the ratings? If they asked the people who keep the apps, just like they do to the people who delete it, then it wouldn't be skewed.

 

[Patrick J]

Youtube recently introduced Like/Dislike as a system for rating videos. With the older 1 to 5 star system, 95% of votes were either 1 or 5.

I wonder if this is happening with the Apple rating system?

 

[redkamel]

Why don't you tell us why? I've been jailbroken for over a year and I love it.

Its been a long time since I've been jailbroken, so some of these may be wrong. Feel free to correct me. Keep in mind this is for MOST PEOPLE

1. It involves using the computer to do something "you are not supposed to do". So its a little scary, and a little confusing, even with step by steps.
2. It may brick their phone with an update, in which case they will be mad.
3. It requires them to use something besides app store to get software on there. So that can be confusing.
4. Some people may not even understand the concept. I have many friends who don't even get how a torrent works.
5. At some point they may find their warranty voided and get mad. Again, many people are not "computer smart" enough (or whatever) to remember to/thin of returning their phone to the original state before going in for genius bar
6. It may require multiple full resets to factory settings which is frustrating.

I personally stopped jailbreaking cause I was over spending 20 minutes figuring out if I should update my iphone just so I could keep a wallpaper. I didn't use multitasking or anything else.

 

[Davey782]

I liked the option to rate on delete. It's no secret that people complain about negative attributes more often than they promote quality. We're complainers.. this is nothing new..

 

[arn]

You are skewing logic here...

If people delete the app as you say most likely don't like it, so they should be able to give it a low rating. Why is this skewing towards low ratings?

you are wrong, it does skew it towards low ratings. this introduces sampling bias to the ratings. Now, you may be ok with that, but it does mean the ratings are lower than if they were from a random sampling of people.

arn

 

[Mr. Zorg]

You are skewing logic here...

If people delete the app as you say most likely don't like it, so they should be able to give it a low rating. Why is this skewing towards low ratings? The apps that are liked and not deleted, don't get bad ratings, but the ones that are deleted, get the bad ratings. If these apps were any good, they would not be deleted and they wouldn't be getting bad ratings? So, ultimately the good apps that are not deleted get the better grades, and those that do get deleted the worst.

I think this not only does not skew ratings, it rectifies them.

I strongly disagree, and I think most developers would too. There are always going to be people who don't like an app for one reason or another, even if the app is perfectly fine for most people.

So what happens is that, say, 10,000 people download an app. Now let's say 250 of those people just can't stand it. Maybe they didn't like the color palette the developer used, maybe they just never really "got" the purpose of the app - so they delete it. They now get prompted for a rating and give it a bad one because of their opinions. As a result, the app gets a rating of 1.5 stars with 250 reviews. Looks like a horrible app, right? Not so. The other 9,750 users like it just fine, but never give it a rating because they didn't get prompted to do so. So you've got an apps rating being determined by a mere 2.5% of it's user base.

How does that rectify the ratings? Either make ratings simple for both kinds of users, or make it hard for both. But making it easy for one and hard for the other is definitely skewing results.

 

[Bytor65]

you are wrong, it does skew it towards low ratings. this introduces sampling bias to the ratings.

In what way? If it skews down ratings on Apps that get deleted a lot, that IMO is a good thing. They were likely deleted because they weren't very good.

From what I read, one of the biggest problems with the Appstore is that it is full of junk Apps. I am for anything that hightlights the junk so I can avoid it.

 

[Sambo110]

I was just thinking about this last night. A free Sudoku app had a 2 and a half star rating with 530 ratings, but the written ratings were all 5 stars. So people download, play it a bit, don't like it then rate it lowly which makes other people skip it because of it.

 

[Mr. Zorg]

Its been a long time since I've been jailbroken, so some of these may be wrong. Feel free to correct me. Keep in mind this is for MOST PEOPLE
...snip...

I'm constantly shocked by how many people I know who have an iPhone and have NEVER even plugged it in to a computer. Not once. Incredible.

 

[MySchizobuddy]

Why don't you tell us why? I've been jailbroken for over a year and I love it.

you cannot install apps from the app store on jail broken phones right?

 

[steamtoy]

Apple should really reconsider their app return policy!

Apple should really reconsider their app return policy! They should at least allow 24 hours for full return. People have wast so much money on apps like they only tried once and never used.

 

[faroZ06]

Spelling Error

It says "as" instead of "ask" in the part about what the iPhone/iPod does when you delete an app.

 

[pubwvj]

Statistically a Bad Idea

Good. I'm glad they're fixing this. From a statistics point of view it is highly biased. Rating should be done at any time, but not at delete. That produces a self-selected negative biased set.

 

[BJWanlund]

Just FYI Arn, you're actually largely ignoring the fact that us USERS actually might like that. I've had to delete MANY an app from my iPhone, and it absolutely drives me insane to have to deal with that ratings screen EVERY TIME I deleted an app. Good move for both developers AND users, Apple.

BJ

 

[Shasterball]

Ya. I actually really agreed with the developers here....

 

[applesupergeek]

It skews the ratings because it makes it more likely for people who dislike (or even marginally dislike) the app to rate it.

In long terms:
it skews the ratings because you give more of opportunity to rate from one population (people who delete the app) than another (people who don't delete the app). These populations are not the same; one is more likely to dislike the app right? They deleted it. So it is makes it more likely for negative reviews to be turned in than positive ones.

Its like conducting a survey of a movie by specifically targeting all the people who walk out in the middle, then leaving the rest of the surveys for whoever wants to pick one up on the way out.

You are completely wrong.

Let me spell it out for you. That's not skewing. It does decrease overall ratings, but it's not skewing them towards a bias of any kind. On the contrary it gives the user a chance to easily rate the bad apps/the unwanted apps when they delete them. So, again, more negative reviews are turned in possibly, but these negative reviews go to deservedly deleted apps.

On the contrary what is skewing the rating unfairly is not giving an easy option to rate a deleted app, because the bad apps don't get a bad rating even though they had to be deleted, and might thus keep a misleading high score similar to the apps that aren't deleted.

Why should apps that are seldom deleted be on a level playing field with those often deleted, why take away the option to rate down the apps that do get deleted for some reason, and thus (whilst as a byproduct bringing ratings uniformly down) enable the often used, kept apps, to have a higher rating?

It's really not that hard to grasp. But of course developers who make crap apps have everything to lose here and they don't want people rating them down when they delete their junk. And they are very vocal here, to the extent that the main article makes the stupid and silly assertion that somehow this skews ratings, which is very unlike arn (I thought initially it was written by Slivka)

 

[*LTD*]

You are completely wrong.

Let me spell it out for you. That's not skewing. It does decrease overall ratings, but it's not skewing them towards a bias of any kind. On the contrary it gives the user a chance to easily rate the bad apps/the unwanted apps when they delete them. So, again, more negative reviews are turned in possibly, but these negative reviews go to deservedly deleted apps.

On the contrary what is skewing the rating unfairly is not giving an easy option to rate a deleted app, because the bad apps don't get a bad rating even though they had to be deleted, and might thus keep a misleading high score similar to the apps that aren't deleted.

Why should apps that are seldom deleted be on a level playing field with those often deleted, why take away the option to rate down the apps that do get deleted for some reason, and thus (whilst as a byproduct bringing ratings uniformly down) enable the often used, kept apps, to have a higher rating?

It's really not that hard to grasp. But of course developers who make crap apps have everything to lose here and they don't want people rating them down when they delete their junk. And they are very vocal here, to the extent that the main article makes the stupid and silly assertion that somehow this skews ratings, which is very unlike arn (I thought initially it was written by Slivka)

http://en.wikipedia.org/wiki/Sampling_bias

 

[Porco]

Great change. I disliked that annoying screen popping up. If I want to rate it I will, don't shove it in my face.

I completely agree. Rating apps should be something the user does of their own volition (whether positively or negatively), not something they are prompted / nagged to do by the device. But then Apple could have just included the rate upon deletion pop-up as an option in the device settings from the beginning instead of deciding for us and then changing their mind.

 

[applesupergeek]

you are wrong, it does skew it towards low ratings. this introduces sampling bias to the ratings. Now, you may be ok with that, but it does mean the ratings are lower than if they were from a random sampling of people.

arn

I am sorry but this is incorrect.

A sampling bias is

causing some members of the population to be less likely to be included than others

. What members of the population are not included here? People who never delete apps? People who delete apps, less often? Everyone deletes apps now and then, and to some extent they do it proportionally to the amount of apps they download? So, who is sampled out here? No one.

I can argue that it's without this system (rating on delete) that a sample bias is introduced: Everyone deletes apps, so no sample bias here, but not a that many people feel compelled to go to the app store and submit a rating. That's the real bias.

The ratings are lower, not because it's not a random sample of people (which it is, no one is selectively chosen, every person who deletes an app has a choice, every person by and large deletes an app) but because negative rating on delete is encouraged.

But as I have argued that not only does not skew the ratings, it actually rectifies them, by urging users to (implicitly) vote down the bad apps that they delete. Hence the total score of apps falls overall, but comparatively the good ones you never delete don't suffer the bad ratings, and can be separated from the bad ones.

The good apps also don't get good ratings for being on the device. The user is never prompted to rate an app if it is left on the device. How is that not skewing the ratings? If they asked the people who keep the apps, just like they do to the people who delete it, then it wouldn't be skewed.

There are no separate groups of people, some that delete, and others that keep apps. Period. Everyone deletes an app now and then, and they are given the option to rate (down). The apps that stay don't get rated down, so how do they suffer? They don't. They actually gain from the bad apps being rated down.

 

[kresh]

http://en.wikipedia.org/wiki/Sampling_bias

Is the point of the rating system to be statistically accurate, or to highlight good apps by having few negative ratings and bad apps with a slew of negative ratings?

To heck with statistical accuracy, I'm all for highlighting crap apps with an over-abundance of appropriately negative ratings.

 

[Nuvi]

The fact is that vast majority of apps getting low reviews deserve them. Ultimately all the great apps have great reviews. If you can't write a good app then eventually you will get bad ratings anyway. Only this time you'll get few words elaborating on that. Honestly what Apple should have done is to have a pop up on after certain amount of app launches so there would have been easy and fast way of giving good and bad news for the devs. Instead of removing "rate on delete" Apple should have been looking ways of improving the system and not killing it. Hell, when you have 100 000 apps you need good a fast way of sorting out the bad seeds.

 

[applesupergeek]

http://en.wikipedia.org/wiki/Sampling_bias

I am sorry Quadra but as per wiki:

Self-selection bias, which is possible whenever the group of people being studied has any form of control over whether to participate. Participants' decision to participate may be correlated with traits that affect the study, making the participants a non-representative sample. For example, people who have strong opinions or substantial knowledge may be more willing to spend time answering a survey than those who do not. Another example is online and phone-in polls, which are biased samples because the respondents are self-selected. Those individuals who are highly motivated to respond, typically individuals who have strong opinions, are overrepresented, and individuals that are indifferent or apathetic are less likely to respond. This often leads to a polarization of responses with extreme perspectives being given a disproportionate weight in the summary. As a result, these types of polls are regarded as unscientific.

As per my italics a sampling bias already exists in that those people are the ones who usually rate in the app store. On the contrary uniformly offered a rating upon delete, suffer much less from this bias than those who have to out of their way and rate in the app store. Think about it, and don't insist on this point, because I don't think you 've grasped what I am saying.