TrueDelta reliability survey results - Subaru Legacy and Outback

[mkaresh]

The latest TrueDelta Vehicle Reliability Survey results have been posted. These are for the year ending June 30, 2007. Late responses could lead to further changes, but these are unlikely to be large.

 

For the 2005 Subaru Legacy and Outback, I have an average of 8.7 months of data on 35 cars. The stat of 39 successful repair trips per 100 vehicles per year (0.4 per vehicle) is better than average for a 2005 model.

 

For the 2006 Subaru Legacy and Outback, I have an average of 5.9 months of data on 21 cars. This is an insufficient sample size, so the result is asterisked and only visible to site members. Anyway, the stat of 10 successful repair trips per 100 vehicles per year (0.1 per vehicle) is so low that I expect it to increase with a larger sample size in future updates. For now, it can only be said that the repair rate is better than average, and probably better than for the 2005s (as it should be).

 

Repairs reported for the 2005 have included:

--squeaking seat frame

--headlight bulbs

--wheel bearings

--warped armrest

--stress crack in windshield

--fan relay (recall, but damaged before recall performed; usually recalls aren't included)

--passenger seat sensor

 

Everyone who has been helping out with this research, thanks, I literally couldn't have done it without you. Those who aren't yet helping out, I hope you'll consider doing so.

 

Other results:

 

TrueDelta Vehicle Reliability Survey results

[rjundi]

Seems like a small data sample but great to capture the data. Out of curiosity do you know what Consumer Reports considers a data sample?

[mkaresh]

Their minimum sample size is 100, but then with that they provide system level scores (engine, transmission, A/C, brakes) that can differ by only a single percentage point. Which you really can't do with a sample that small.

 

For a fairly accurate (within 15-20 percentage points) estimation of overall repair rates, 25 is much more adequate than 100 is for what they do.

 

I'm only including the things that people reported problems with because many people like to see this information. My sample size is not large enough to conclude that these are common problem areas.

 

I'll be providing additional, more detailed results in the future, when the sample sizes are larger. I'm focusing on the overall repair trip rate right now because it can be estimated with relatively small sample sizes.