The average reading performance of grade 6 pupils was found to be 80 words per minute. A random sample of 20 students from three sections took the reading test and found to have an average speed of 74 words with a standard deviation of 8 words per minute. Can we say that the students performance is below the standard when tested at 0.05 level of significance?
Since the assumption that student's performance is below the standard is equivalent to assumption that population standard is below 80(in terms of testing), then
"H_0:a=80"
"H_1:a<80" , where a - population mean
Since the population standard deviation is unknown and sample size is small then we should run t-test
t-statistic is calculated by next formula
"t={\\frac {x-a} {\\sigma}}*\\sqrt{n}" , where x - sample mean, a - supposed population mean, n - sample sizq, "\\sigma" - sample standard deviation
In the given case we have
"t={\\frac {74-80} {8}}*\\sqrt{20}=-3.35"
According to the form of alternative hypothesis, we will have enough statistical evidence to reject null hypothesis if t < Cr, where Cr is such value, that
"P(T(n-1)<Cr)=a=0.05" , where T - Studnent's criteria with n - 1 degrees of freedom, n - sample size, a - level of significance. In the given case we have
"P(T(19)<Cr)=0.05\\implies \u0421r = -1.73"
We receive that t < Cr, so we should reject the null hypothesis and admit that student's performance was indeed below standard at 0.05 level of significance
Comments
Leave a comment