Which statistical procedure to use (given standard deviations and means)?

statstand1991

New member
Joined
May 8, 2013
Messages
1
Can someone please help me with this? Which statistical procedure could I use, and how could I set this up?

20 males and 20 females reported the number of traffic violations they received in the last 8 years.
Females: 9, 1, 2, 4, 4, 3, 8, 8, 3, 2, 4, 8, 6, 7, 6, 5, 5, 5, 5, 5
Males: 10, 4, 6, 9, 8, 5, 5, 9, 3, 8, 8, 5, 3, 8, 7, 6, 7, 4, 8

The DMV is concerned that there has been an increase in traffic tickets in the last 8 years. Based on data collected continuously in the 8 years PRIOR to when the above data was collected, the average number of tickets was 2 with a standard deviation of 1. Using the data for both males and females combined (ignore gender), has there been a significant increase in the number of traffic tickets?

I was thinking maybe a one-sample z test, but I'm unsure because I do not know if the data for 8 years prior follows a normal distribution or what the N is.

Thank you.
 
Can someone please help me with this? Which statistical procedure could I use, and how could I set this up?

20 males and 20 females reported the number of traffic violations they received in the last 8 years.
Females: 9, 1, 2, 4, 4, 3, 8, 8, 3, 2, 4, 8, 6, 7, 6, 5, 5, 5, 5, 5
Males: 10, 4, 6, 9, 8, 5, 5, 9, 3, 8, 8, 5, 3, 8, 7, 6, 7, 4, 8

The DMV is concerned that there has been an increase in traffic tickets in the last 8 years. Based on data collected continuously in the 8 years PRIOR to when the above data was collected, the average number of tickets was 2 with a standard deviation of 1. Using the data for both males and females combined (ignore gender), has there been a significant increase in the number of traffic tickets?

I was thinking maybe a one-sample z test, but I'm unsure because I do not know if the data for 8 years prior follows a normal distribution or what the N is.

Thank you.
Please check the question - the 40 recent data are so different from the stated mean that it is hardly worthwhile even to do a statistical test. If the null hypothesis is that the data are a sample from a distribution with mean = 2, only 3 of the 40 are less than or equal to the mean!

Perhaps it said (or should have said) the preceding mean was 5? That would at least make the question interesting.

You don't know the shape of the parent distribution for H0 - just the mean and standard deviation. For a sample of size N=40, the distribution of sample means will be normal with the same mean as the population, and the standard deviation will be divided by sqrt(40). I would just use the sample mean as the test statistic, comparing it to the normal distribution.

Show us more of your work!
 
Top