In: Statistics and Probability
Hepatitis C is a blood-borne infection with potentially serious
consequences. Identification of social and environmental risk
factors is important because Hepatitis C can go undetected for
years after infection. A study conducted in Texas in 1991-2
examined whether the incidence of hepatitis C was related to
whether people had tattoos and where they obtained their tattoos.
Data were obtained from existing medical records of patients who
were being treated for conditions that were not blood-related
disorders. The patients were classified according to hepatitis C
status (whether they had it or not) and tattoo status (tattoo from
tattoo parlour, tattoo obtained elsewhere, or no tattoo). The data
are summarised in the following table.
Tattoo? Has Hep C No Hep C Tattoo (parlour) 17 43 Tattoo
(elsewhere) 8 54 No tattoo 22 461 (a) In any association between
hepatitis C status and tattoo status, which variable would be the
explanatory variable? Justify your answer. [2] (b) If a simple
random sample is not available, a sample may be treated as if it
was randomly selected provided that the sampling process was
unbiased with respect to the research question. On the information
provided above, and for the purposes of investigating a possible
relation between tattoos and hepatitis C, is it reasonable to treat
the data as if it was randomly selected? Briefly discuss. [2] (c)
Assuming that any concerns about data collection can be resolved,
evaluate the evidence that hepatitis C status and tattoo status are
related in the relevant population. If you conclude that there is a
relationship, describe it. Use a 1% significance level.
Observed = Oi | Hep C | No Hep C | Total |
Tattoo (parlour) | 17 | 43 | 60 |
Tattoo (elsewhere) | 8 | 54 | 62 |
No tattoo | 22 | 461 | 483 |
totals | 47 | 558 | 605 |
Expected frequency of a cell = sum of row*sum of column / total sum
Expected | Hep C | No Hep C |
Tattoo (parlour) | 47*60/605= 4.66 | 558*60/605= 55.34 |
Tattoo (elsewhere) | 47*62/605= 4.82 | 558*62/605= 57.18 |
No tattoo | 47*483/605= 37.52 | 558*483/605= 445.48 |
Expected Count = Ei | Hep C | No Hep C |
Tattoo (parlour) | 4.66 | 55.34 |
Tattoo (elsewhere) | 4.82 | 57.18 |
No tattoo | 37.52 | 445.48 |
Chi square =(Oi-Ei)^2/Ei | Hep C | No Hep C |
Tattoo (parlour) | 32.68 | 2.75 |
Tattoo (elsewhere) | 2.1 | 0.18 |
No tattoo | 6.42 | 0.54 |
44.67 |
Chi-Square Test Statistic,χ² = Σ(fo-fe)^2/fe =
44.658
Level of Significance = 0.01
Number of Rows = 3
Number of Columns = 2
Degrees of Freedom=(#row - 1)(#column -1) = (3- 1 ) * ( 2- 1 )
= 2
p-Value = 0.0000 [Excel function:
=CHISQ.DIST.RT(χ²,df) ]
Decision: p-value < α , Reject Ho
It can be concluded that there is an association between two variables.
Hepatitis status---Explanatory
b)
We can data is reasonable unbiased as expected frequency is greater than 5
c)
p-Value = 0.0000 [Excel function:
=CHISQ.DIST.RT(χ²,df) ]
Decision: p-value < α , Reject Ho
======================================================
If you need any clarification please let me know through
comment
Please give positive vote if you find this
solution helpful. Thank you!