Question

I'm doing a stats project on the correlation between the amount of hours a person sleeps and their weight; proving there is no correlation between sleep and weight gain. I have voluntary samples of people's weight and the hours they slept, and I need to list if the level of measurement is nominal or ordinal, confidence interval, and linear correlation test. how do I do this?

Answer 1

Sol:

label x as hours slep

y as weight

perform cor.test in R(which tests whethe the relationship is significant or not)

if p>0.05

relationship i s significant.

ex: i am taking hours as

2,3,4,5,7

weight as

54,45,67,78.34

my null hypothesis:

THere is no relationship between hours slept and weight

alternative hypothesis:

There is a negative relationship between hours slept and weight.

Rcode is:

hourssslept <- c(2,3,4,5,7)
weight <- c(54,45,67,78,34)
cor.test(hourssslept,weight)

ouptut:

Pearson's product-moment correlation

data: hourssslept and weight
t = -0.3784, df = 3, p-value = 0.7303
alternative hypothesis: true correlation is not equal to 0
95 percent confidence interval:
-0.9220699 0.8239945
sample estimates:
cor
-0.2134366

here corr coefficient is r=-0.2134366

p=0.7303

p>0.05

Fail to reject H0.

Accept H0

THere is no relationship

to find confidenc einterval in R

Rcode is

t.test(hourssslept,weight)

Welch Two Sample t-test

data: hourssslept and weight

t = -6.5596, df = 4.0976, p-value = 0.002567

alternative hypothesis: true difference in means is not equal to 0

95 percent confidence interval 95 percent confidence interval::

-72.9529 -29.8471

sample estimates:

mean of x mean of y

4.2 55.6

output:

95 percent confidence interval: s from 4.2 to 55.6