Slashdot Mirror
Apple Laptop Reliability Survey
Nikopol writes "The venerable Macintouch site recently released the results and analysis of a survey on Apple laptop reliability." From the article: "Our survey spanned every laptop capable of running Mac OS X, encompassing 41 models sold over seven and a half years. A challenge in ascertaining the reliability of any device is that more time gives them more opportunity to break, so new devices should always look more reliable at first glance. Our survey asked participants when their laptop first needed a repair -- 'first year', 'second or third year', and 'fourth year or later.' These correspond to the duration of Apple's standard one-year warranty, the AppleCare extended warranty program, and any repairs that might happen outside any warranty coverage period. We also asked participants if they purchased AppleCare for their laptop."
As much as people like to quote the macintouch surveys, they are still self-selected unverified Internet studies. As such, they are not useful for any real-world decision making, any more than Slashdot polls.
Macintouch claims that this is not a problem, but they have no way to support that claim.
I completely agree. This is a meaningless poll, something I've written to Ric Ford about a couple of times, without reply. The Macintouch folks don't seem to understand how to conduct a meaningful survey and don't seem to care that their results are likely misleading. To be accurate, we have no way of knowing if their result are meaningful or not, since they have no way of measuring the biases in their sampling.
A pity this kind of nonsense is published by Macintouch and an equal pity that a site that claims to be "news for nerds" propagates an unscientific poll like this.
Sailing over the event horizon
Introduction
We thank MacInTouch readers who participated in our recent laptop survey. More than 10,000 laptops were logged, along with many thousands of comments.
Survey Design
Our survey spanned every laptop capable of running Mac OS X, encompassing 41 models sold over seven and a half years. A challenge in ascertaining the reliability of any device is that more time gives them more opportunity to break, so new devices should always look more reliable at first glance. Our survey asked participants when their laptop first needed a repair -- "first year", "second or third year", and "fourth year or later." These correspond to the duration of Apple's standard one-year warranty, the AppleCare extended warranty program, and any repairs that might happen outside any warranty coverage period.
We also asked participants if they purchased AppleCare for their laptop. MacInTouch readers have strong opinions about whether or not AppleCare is a good investment; we hoped to quantify how often it is invoked for service. Hoping to gauge which models might have had repeat problems, we also asked how many times a laptop was repaired (regardless of warranty status) -- as well as providing an option for "It broke, but I didn't have it fixed." (Thus, repair numbers do not always exactly match problem reports.)
We also asked what components were repaired or replaced (if known), with check boxes for most major laptop components, as well as an "other" choice.
Finally, since accidents happen, we asked if the laptop was ever dropped.
Summary
* Overall average: 41% repaired (lifetime)
* Overall standard deviation: 0.17 (large variation between models)
Fewest Repairs:
* original (colored) iBooks
* PowerBook G3
* 12" PowerBook G4 (2005)
* 12" iBook G4 (recent models)
Most Repairs:
* iBook G3 (esp. in 2003, motherboard defects)
* 15" titanium PowerBook G4 (esp. in 2003, optical drive, case/latch/hinge)
* 15" aluminum PowerBook G4 (first version, screen defects)
Component Failures
* Overall component failure average: 5%
Standard deviation: 0.035 (large variation between component types)
* Most common failures (>1 STDEV above average): motherboard, display, hard drive
* Least common failures (>1 STDEV below average): RAM
* Less common failures (0.5-1 STDEV below average): keyboard, trackpad, case, latch
Charts
Repairs by Model and Time
Green cells indicate repairs significantly below average (ie, more reliable)
Red cells indicate repairs significantly above average (ie, less reliable)
iBook and PowerBook repairs by model
* Numbers may not add up to 100% due to rounding or variations in owner reporting.
Some laptops failed but were not repaired, so total failures (43%) are slightly higher than total reported repairs (40.6%).
Component Repairs
Green cells indicate repairs significantly below average (ie, more reliable)
Red cells indicate repairs significantly above average (ie, less reliable)
Table of Components Repairs
(click to enlarge)
* Numbers may not add up to 100% due to rounding or variations in owner reporting.
Analysis
In our analysis, we focus largely on first-year and second- and third-year repairs. Although we tracked fourth-year and later repairs, mostly it just tells us older models tend to break down.
The Titanium PowerBook G4s, which were quite popular, were more broadly troublesome than most models, with optical drive problems and case/latch/hinge problems being the most common issues.
But by affected