Andrey works at a call center, selling insurance over the phone. He usually starts his calls with "Hello!" but he thinks that starting with "Howdy!" might improve the prospects of a juicy sale.
He randomized his subsequent $500$ calls between a treatment group and a control group. He used "Hello!" in each of the control group calls and "Howdy!" in each of the treatment group calls, and he kept track of the sale price he achieved in each call.
The results of the experiment showed that the mean sale price of the treatment group was $\$20$ more than the mean of the control group. To test whether the results could be explained by random chance, Andrey created the table below, which summarizes the results of $1000$ re-randomizations of the data (with differences between means rounded to the nearest $10$ dollars).
**According to the simulations, what is the probability of the treatment group's mean being higher than the control group's mean by $\$20$ or more?**
$\qquad$[[☃ input-number 1]]
Assume that if the probability you found is *lower* than $5\%$, then the result should be considered significant.
**What should we conclude regarding the experiment's result?**
[[☃ radio 1]]
Treatment group mean $-$ Control group mean | Frequency
:-: | :-:
$-50$ | $7$
$-40$ | $13$
$-30$ | $65$
$-20$ | $123$
$-10$ | $189$
$0$ | $233$
$10$ | $159$
$20$ | $125$
$30$ | $67$
$40$ | $14$
$50$ | $5$