Question

Should statistical analyses, i.e., confidence intervals for the population mean and hypotheses testing about the population mean, be conducted using the data on the original or on the log-transformed scale?

The original data is not normal and the log-transformed scale is normal.

Answer 1

The log-change is broadly utilized in biomedical and psychosocial research to manage skewed information.

This paper features difficult issues in this exemplary methodology for managing skewed information.

In spite of the regular conviction that the log change can diminish the changeability of information and influence information to adjust all the more intently to the typical appropriation, this is normally not the situation.

Additionally, the consequences of standard factual tests performed on log-changed information are regularly not significant for the first, non-changed information.

The log transformation, a widely used method to address skewed data, is one of the most popular transformations used in biomedical and psychosocial research.

Due to its ease of use and popularity, the log transformation is included in most major statistical software packages including SAS, Splus and SPSS.

In any case, when you utilize a factual model, you are managing how this model sees the world and its presumptions for exact execution, not simply the universe of estimations.

Therefore, tossing in an exceptionally skewed variable in direct relapse, is requesting that it accomplish something it can't: to convey exact gauges within the sight of anomalies, heterogeneity of change and absence of multivariate ordinariness

. Strategic relapse is in reality less touchy to distributional issues, however is delicate to exceptions.

In the event that you plot a variable against its log, or some other power-change, you will see that the connection between them is non-straight. In this way, when you change a variable to its log, you are all the time linearizing the non-direct connection between the result and your autonomous variable.

Straight relations are more reproducible and understandible, despite the fact that few are not (maybe most), than say, composing your own non-direct capacity.

You can without much of a stretch check this by plotting the untransformed and changed adaptations against the result.