**Analysis**

**Descriptive statistics split by condition**

**Control condition [Prime code = 0]**

```
## vars n mean sd min max range se
## X1 1 57 5.83 1.76 1.5 9 7.5 0.23
```

**Aggression condition [Prime code = 1]**

```
## vars n mean sd min max range se
## X1 1 59 5.44 1.93 2 9 7 0.25
```

**Independent samples ***t*-test with mean aggression rating as *DV* and prime condition as *IV*

Gives ordinary asymptotic *p*-value.

```
##
## Welch Two Sample t-test
##
## data: dv by prime
## t = 1.1453, df = 113.6, p-value = 0.2545
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -0.286506 1.071817
## sample estimates:
## mean in group 0 mean in group 1
## 5.833333 5.440678
```

**Cohenâ€™s ***d* effect size

```
##
## Cohen's d
##
## d estimate: 0.2123703 (small)
## 95 percent confidence interval:
## inf sup
## -0.1598214 0.5845620
```

**Exact two-tailed ***p*-value and associated 99%*CI* for the *p*-value from a Monte-Carlo permutation (randomization) test

Apart from precision, permutation test sidesteps conditioning on any distributional assumptions.

```
##
## Exact Permutation Test Estimated by Monte Carlo
##
## data: GROUP 1 and GROUP 2
## p-value = 0.2608
## alternative hypothesis: true mean GROUP 1 - mean GROUP 2 is 0
## sample estimates:
## mean GROUP 1 - mean GROUP 2
## 0.3926554
##
## p-value estimated from 1e+05 Monte Carlo replications
## 99 percent confidence interval on p-value:
## 0.2572698 0.2644330
```

**Bayesian analysis**

**Compute Bayes Factor for independent ***t*-test of a one-tailed hypothesis

For the purpose of quantifying the available evidence. The Bayes factor is the relative evidence in the data. The evidence in the data favors one hypothesis, relative to another, exactly to the degree that the hypothesis predicts the observed data better than the other (answering the question how many times are the data more likely under one hypothesis as oposed to the other. BF10 = evidence in favor of *Ha*, BF01 = evidence in favor of *H0*. More info here. Using a prior width of 1/2. Testing a directional, one-tailed hypothesis that aggressive priming leads to a higher rating of aggression.

`## [1] "BF01 = 7.28"`

**Bayesian sequential analysis**

Plotted evolving empirical support (relative likelihood) in favor of *Ha* or *H0* as more and more data come in. More info here.

`## NULL`