# How do I compare categorical data with multiple uneven populations?

Is there a difference in lineage occurrence/proportion between study 1 compared to the other 4 studies?

There are 5 populations. We need to compare population 1 to each of the other 4 populations (4 previous studies).

We are comparing the occurrence of Lineages (4 Levels- A, B, C, D) in population 1 compared to their occurrence in each of the other 4 populations.

The Lineages are categorical (A,B,C, D) every sample will fit within one of these lineages. The sample sizes are uneven between the different studies, therefore I thought it best to work with proportions.

I thought that I could just use a proportion test or chi-square test but I'm a bit confused since I have the 4 lineages and the 5 studies. Do I just look at one lineage at a time and since it's a proportion, it factors in the other lineages?

i.e.: LineageB <- prop.test(x = c(198,140,31,35,205), n = c(213,140,31,39,250)); LineageB

p-value = 3.502e-08 (This isn't specific enough, tells me there is likely a different between all studies)

Since I'm only concerned with population 1 compared to each of the others should I only include my study and one other at a time?

Lineages | Pop 1 |
Pop 2 |
Pop 3 |
Pop 4 |
Pop 5 |

A | 4 | 0 | 0 | 0 | 0 |

B | 198 | 140 | 31 | 35 | 205 |

C | 6 | 0 | 0 | 4 | 18 |

D | 5 | 0 | 0 | 0 | 27 |

Total |
213 | 140 | 31 | 39 | 250 |

## Answer

**Answers can be viewed only if**

- The questioner was satisfied and accepted the answer, or
- The answer was disputed, but the judge evaluated it as 100% correct.

- answered
- 195 views
- $6.00

### Related Questions

- Statistical significance
- Define$ F : C[0, 1] → C[0, 1] by F(f) = f^2$. For each $p, q ∈ \{1, 2, ∞\}$, determine whether $F : (C[0, 1], d_p) → (C[0, 1], d_q)$ is continuous
- $\textbf{I would like a proof in detail of the following question.}$
- elements in a set
- Populace Model
- Prove that every compact Hausdorff space is normal
- Quantitative reasoning
- Correlation of Normal Random Variables