I am trying to figure out what type of statistical test to u

I am trying to figure out what type of statistical test to use to analyse my data from an independent experiment. we tested if female bean beatles prefered a used bean to lay eggs on or a new clean bean so we placed one female with both type to choose from. we collected how many eggs new eggs were layed on each and I have calculated the average for both types. I was thinking a t-test but i\'m not sure if that would be the right type to see if the results were significant. what would be the right type of test to do and if there is a good link for an online calculator. Now I just asked this and got answer about t-test could work, but there is different types of t-test so which one is right one ? they also mentioned a chi squ test but we didn\'t have expected value so maybe if t-test will work please follow up on the type of one .

Solution

Ans. Your data includes-

            I. Two groups – 1. Used beans    , 2. Clean beans

            II. Quantitative data- number of eggs laid in each group

You want to test – “if the female preferred ‘used’ or ‘clean’ beans to lay egg BY counting the number of eggs in each group”.

Aim of the Statistics:

            I. To test if the two groups are same or different in terms of number of eggs.

Possible statistical strategies:

1. t-test: t-test is preferred when the two sets of data are result of independent observations and follow normal distribution. It compares the mean of two or more populations.

t-test is further divided into parametric and non-parametric t-test. Parametric t-test is conducted when the arithmetic mean accurately represents the centre of distribution of a sufficiently large population size. Non-parametric t-test is conducted when the median accurately represents the centre of distribution. If your sample size is large, then go for parametric t-test. For a small sample size, go for non-parametric t-test.

2. X² test- X²is preferred for categorical data, i.e. each group is divided into two or more categories, say, red and white color of flowers.

Since, the data is categorical- (used vs clean bean groups), chi-square test shall be preferred.

Expected values are required to compare a group data (observed) with a pre-standardized data (expected).

Since, here you have to compare two groups (category), you place the two groups in two columns and their respective values in corresponding rows. There is no need of ‘expected values’ while comparing the two groups here.

Online calculator: http://turner.faculty.swau.edu/mathematics/math241/materials/contablecalc/