Question

When would you use ANOVA instead of a chi-square test?

Is it possible to establish through a regression analysis if a variable is a dependent or an independent variable? Why or why not?

Answer 1

- ANOVA represents examination of fluctuation. It thinks about at least 3 methods dependent on at least 1 elements. It does this by contrasting the difference inside gatherings with the change between gatherings. On the off chance that the last is a lot bigger than the previous, at that point you reason that there is a distinction in implies between gatherings. It is identical to straight relapse.

ANOVA or Analysis of Variance is a method that is utilized to look at various methods. The strategy examinations the analyzed methods dependent on between and inside gathering fluctuation.

The aftereffect of ANOVA just tells that at any rate one of the thought about methods is extraordinary and not which one of them.

Chi square is utilized for an assortment of things. There are single direction, two-way and more than two-way chi-square tests. In any case, they share for all intents and purpose that all the factors are clear cut and that none is a needy variable. In the event that you have an all out ward variable you will presumably need some type of strategic relapse.

A chi-square test is for the most part used to discover the "integrity of fit" or to "test autonomy" of clear cut factors. If there should arise an occurrence of an "integrity of fit" test the point of chi-square test is check whether the variable fits the accepted dissemination or not (Observed - expected rationale on the off chance that you may). If there should arise an occurrence of trial of freedom the point is to test if two unmitigated factors are autonomous are most certainly not.

- Regression analysis is a related system to evaluate the connection between a result variable and at least one hazard factors or puzzling factors (bewildering is talked about later). The result variable is additionally called the reaction or ward variable, and the hazard elements and confounders are known as the indicators, or informative or free factors. In regression analysis, the reliant variable is signified "Y" and the free factors are meant by "X".

Basic direct regression is a procedure that is suitable to comprehend the relationship between one free (or indicator) variable and one consistent ward (or result) variable.

Reasonably, on the off chance that the estimations of X gave an ideal forecast of Y, at that point the aggregate of the squared contrasts among watched and anticipated estimations of Y would be 0. That would imply that changeability in Y could be totally clarified by contrasts in X. Be that as it may, in the event that the contrasts among watched and anticipated qualities are not 0, at that point we can't totally represent contrasts in Y dependent on X, at that point there are remaining blunders in the expectation. The lingering mistake could result from incorrect estimations of X or Y, or there could be different factors other than X that influence the estimation of Y.

Connection and direct regression analysis are factual strategies to measure relationship between a free, some of the time called an indicator, variable (X) and a nonstop needy result variable (Y). For relationship analysis, the autonomous variable (X) can be nonstop (e.g., gestational age) or ordinal (e.g., expanding classes of cigarettes every day). Regression analysis can likewise oblige dichotomous free factors.

The methodology depicted here expect that the relationship between the free and ward factors is straight. With certain alterations, regression analysis can likewise be utilized to evaluate affiliations that follow another practical structure (e.g., curvilinear, quadratic). Here we think about relationship between one autonomous variable and one consistent ward variable. The regression analysis is called basic direct regression - straightforward right now to the way that there is a solitary free factor. In the following module, we consider regression analysis with a few free factors, or indicators, considered at the same time.