dondonbabyraptor
Well-Known Member
Awesome, agreed on all your points. Nice to see another life science field. You definitely speak like you have research methods under your belt haha. And sorry you’re definitely right about the study of fuel dilution in relation to certain variable. I didn’t word my point correctly that we are missing possible variables the general user’s may factor in, like you mentioned.About 2/3 of the respondents have reported some sort of mod(s), although that includes people who only reported installing a tailgate damper.
This is coming, once I have a little more time. Tunes are one thing I wish I had asked specifically about, but thankfully quite a few people have volunteered that information.
Life sciences. Not mechanical, but basic analyses like this transcend the scientific fields.
Shouldn't need to be. I'm not trying to assess the fuel dilution issue in relation to its prevalence in the general population, but rather I'm trying to assess fuel dilution as it relates to individual variables. Thus, I need representatives of the different variables, not representatives of the general population. I will grant, however, that the non-random sample that responds to this survey may be essentially missing key variables that are present in the general population.
Yeah, I would say this is the biggest limitation, especially for subgroup and multivariate analyses.
This will be coming
I didn't ask about OCI, but probably should have. I expect, though, that lower OCI would be associated with higher fuel dilution. Not because a shorter OCI would cause fuel dilution, but because folks who know they have fuel dilution will probably be doing more frequent oil changes to mitigate the problem.
I just want to add my humble opinion that we should not got too caught up in "statistical significance." P-values are a fantastic tool (which is why I calculate them), but they are not without their limitations, especially for small sample sizes. I certainly don't advocate for the abolishment of p-values or even for the elimination of significance thresholds, but I do think it important that statistics, like all data, be interpreted within their proper context (including the underlying assumptions of the statistical tests employed). In other words, I do not completely agree with this commentary published in Nature a couple years ago, but I do think its authors raised some important, valid points.
you bring up a great point on p values, it’s definitely a powerful tool and technically any data can be modified/cleaned to have sig p value.But in regards to the data, .7 inches doesn’t seem too big of a difference in an avg n of 25, then again I don’t know much about trucks/cars. Just can’t wait to see more data from here, a little bit of skepticism on the low sample is all. Great stuff and keep rockin’ on with the data collection and great ability of explaining things in laymen terms. Not the easiest thing to do, I see you have some conference or public speaking experience perhaps haha.
Sponsored