Thread: Am I Too Sensitive?

Originally Posted by gusthedog

For any statistics to imply causation we need random sample. In the case of Land Rovers and reliability of new vehicles - you get the case where you can actually track the whole sample under warranty.

I agree with you it is not statistics that is faulty it's the problematic use of statistics that is the issue - the devil is in the details!

I am not so sure you can infer causation with a random sample even with large sample sizes with the above design. I am not sure if you are suggesting that participants each given new vehicles, however even then what would it really be measuring?

Warning: unnecessary experimental design geekery ahead...🤓

From psychological science perspective, to suggest for example x (LR ownership) results in y (+/- satisfaction) you would first need to operationally define dependent variables (DV). For example if a DV was 'recall' what does that even mean for buyers? Why? Because there is no standardised definition of what a recall constitutes, if this is good or bad, what the response to it is, and what event triggers it etc etc - therefore 'recall' is a fairly informationally empty/unhelpful dependent variable.

Next you would need to present independent variables presented to participants randomly allocated to control and experimental groups (preferably with a double blind design i.e., the experimenter and participant not aware of the allocation). Without this causality cannot be inferred.

You would also need control variables for e.g., hours of TV watched weekly, if there was the possibility that exposure to differing advertising rates across brands an influence on participants. You could also try to assess existing biases towards one brand over another that may contaminate data etc etc...

All this doesn't seem particularly practical though...

For a reliability survey to be meaningful every vehicle would have to be driven by the same driver in the same way over the same terrain the same number of times in the same conditions and given the same maintenance. Doesn't happen.

Originally Posted by DiscoMick

For a reliability survey to be meaningful every vehicle would have to be driven by the same driver in the same way over the same terrain the same number of times in the same conditions and given the same maintenance. Doesn't happen.

...and because it doesn't happen, such a survey would be pointless anyway.

I'm interested I where 'character' comes into to the drivers preference for a brand or vehicle model. Character or the particular quality of the experience often outweighs perceived reliability, to the point where even when something mechanical or electrical fails, the owner goes into denial and proclaims their vehicle, whatever brand it is, reliable.

Originally Posted by DiscoMick

For a reliability survey to be meaningful every vehicle would have to be driven by the same driver in the same way over the same terrain the same number of times in the same conditions and given the same maintenance. Doesn't happen.

There is a further problem that not all vehicles are designed for the same role. Even amongst 4x4s, there are some that cope better with some things than others.

A comparison could be biased to favour one that was better at carrying heavy loads by insisting that the test cargo load was very heavy. One that has sturdier suspension would be favoured if the test route comprised badly corrugated roads even though other models might last better with continuous high speed driving on sealed roads.

What does reliability mean anyway? We just reported a broken clip on the parcel shelf in the Mazda. Mazda refused to fix it and wants to charge us BTW. That report doesn't show the car is unreliable though.

I think we're getting a bit too concerned with the detail. A random sample with random use that you could categorise into similar use patterns would be enough. If the sample was big enough it would be ok. Also if the questions asked could only provoke a yes or no answer it would be easier to qualify. For example has your 1999 disco started every time you turned the key? This could then be followed up with a range of choices - 1-2 times, 2-10, 10-50 etc. By compiling a list like this you would of course have variables within in it that you cant qualify. However you would be getting some great data nonetheless.

I'm not talking about a full double-blind scientific randomisation and yes, there would be some error and some multiplying factors such as load etc.

If it was for a TD5 for example though, you could do a yes no for all of the common issues (oil pump bolt, copper washers, sump leaks etc etc). This info taken from a random section of TD5 owners would give you quality data if a little biased. If however you surveyed everyone that had a complaint about their TD5 you would get skewed data. That's where the random selection comes in.

Nothing in the real world is perfect. We're not talking about laboratory conditions here - however you could prove that there was a statistically significant issue regardless of use with the data collection I'm suggesting. It would be better than what we have now anyway

Well My Jan 94 built 200 TDi Discovery is still going strong with over 415000 klms on it. Admittedly I did just have to do a water pump and last year replace the gearbox. But considering it was the first time either had been done EVER. I think it shows just how good these things are. Not to mention the looks you get out bush when you make something look easy that a Cruiser or Patrol struggled with.

Originally Posted by bln

In the last Pat Callinans 4x4 magazine (issue 028) there was a great letter to the editor "Land Rover Loyal". Basically the letter was about a bloke who has been very happy with the many land rovers that he has had in the past. The editor (Scott) responded with something like "its great to hear from a Land Rover Tragic (most of them involving mechanical repairs)...." . Well that's like a red rag to a bull for me. Ive owned 5 land rovers now over the years and I have never had any sort of reliability issues. The odd repairs required from time to time but no more than you'd expect from any other type of 4x4.
Are the 4x4 mags perpetuating an untruth and misrepresenting Land Rover? On what grounds do they claim LR to be unreliable - where is the evidence?. As far as know, and from my experience they are solid reliable 4x4s that excell at there intended use.
I own a small business and I had 3 hiluxs and 2 prados - we had continual problems with them - oil leaks, gearbox overheating, tail shaft faults, recalls as wel l as faulty injectors. These are well known faults so why aren't journalists critical of Toyotas? We have since moved to colorados and our running costs have significantly dropped. I personally traded in my prado for a D4 and the difference is like chalk and cheese.
So my question is - should we be tolerating false journalism or am I just too sensitive? Should Land Rover be tolerating false statements.
I could go on rambling but i look forward to reading what others think.
cheers

To me it reads like the five time LR owner has had a dream run. Not so for me with two out of two failures. First was a 2012 Discovery 4 2.7 litre. It became apparent that over an 18 month period that substantial quantities diesel fuel were entering the sump. At the time of realising the problem, the sump contained two parts oil and one part diesel. The issue was acknowledged by LRA and rectified. However, a vehicle run for an extended period on such diluted engine oil was not for me.

I then bought a new 2015 Discovery 3.0 litre. At 25,000km and after two oil changes at 10k and 23k, the vehicle suffered a catastrophic engine failure in remote Central Australia. The official explanation: the crankshaft dislodged from its housing. This failure occurred while driving on a smooth bitumen surface at 100kph. Needless to say, I was not a happy chappy. However, despite these travails, I still own this second vehicle and hope to for a few more years yet.

My experiences represent significant flaws in the product and in my view cast serious doubt on their reliability. In every other respect they are a fantastic vehicle, particularly off the beaten track.

Nick