Index: The Book of Statistical Proofs ▷ Statistical Models ▷ Count data ▷ Binary contingency table ▷ Definition

Definition: Let $A$ and $B$ be random events and let $\overline{A}$ and $\overline{B}$ denote their complements. Then, (i) the set of relative frequencies of $A$’s and $B$’s occurence or (ii) the set of absolute frequencies of the combinations of $A$/$\overline{A}$ and $B$/$\overline{B}$ in a sample is referred to as a $2 \times 2$ contingency table.

Observed data from a $2 \times 2$ contingency table are characterized as follows:

  • $y_1 = y_{11}$ is the number of cases in which $A$ and $B$;

  • $y_2 = y_{10}$ is the number of cases in which $A$ and $\overline{B}$;

  • $y_3 = y_{01}$ is the number of cases in which $\overline{A}$ and $B$;

  • $y_4 = y_{00}$ is the number of cases in which $\overline{A}$ and $\overline{B}$;

  • $y_{1 \bullet} = y_{11} + y_{10}$ is the (marginal total) number of cases in which $A$;

  • $y_{0 \bullet} = y_{01} + y_{00}$ is the (marginal total) number of cases in which $\overline{A}$;

  • $y_{\bullet 1} = y_{11} + y_{01}$ is the (marginal total) number of cases in which $B$;

  • $y_{\bullet 0} = y_{10} + y_{00}$ is the (marginal total) number of cases in which $\overline{B}$;

  • $n = y_{11} + y_{10} + y_{01} + y_{00}$ is the total number of cases.

The distribution underlying a $2 \times 2$ contingency table is characterized as follows:

  • $p_1 = p_{11}$ is the joint probability that $A$ and $B$;

  • $p_2 = p_{10}$ is the probability that $A$ and $\overline{B}$;

  • $p_3 = p_{01}$ is the probability that $\overline{A}$ and $B$;

  • $p_4 = p_{00}$ is the probability that $\overline{A}$ and $\overline{B}$;

  • $p_{1 \bullet} = p_{11} + p_{10}$ is the marginal probability that $A$;

  • $p_{0 \bullet} = p_{01} + p_{00}$ is the marginal probability that $\overline{A}$;

  • $p_{\bullet 1} = p_{11} + p_{01}$ is the marginal probability that $B$;

  • $p_{\bullet 0} = p_{10} + p_{00}$ is the marginal probability that $\overline{B}$;

  • $1 = p_{11} + p_{10} + p_{01} + p_{00}$ is the probability of the sample space.

 
Sources:

Metadata: ID: D239 | shortcut: ct2x2 | author: JoramSoch | date: 2026-09-11, 09:58.