Suppose you conducted a study to determine if some people have the ability to predict which of two cards a poker player has been dealt for a poker hand. The study involves a single sample of people...

Suppose you conducted a study to determine if some people have the ability to predict which of two cards a poker player has been dealt for a poker hand. The study involves a single sample of people who claim to be “better than average poker players.” In the study, all participants are asked to view a poker player being given a card from the dealer. They are asked to judge from the player’s reaction whether the card is a card dealt to the player that will “complete a four of a kind hand” or “not complete a four of a kind hand.” They complete 100 judgment trials, and their overall percentage accuracy is calculated. In other words, across all trials, the participants have a 50% chance of being correct if they have no ability to interpret the player’s “poker face” (i.e., they are guessing). If they are able to interpret the player’s poker face, the participants should perform better than chance (50%) in correctly choosing which type of card the player has been dealt.

1. Should a one-tailed or two-tailed test be used for this study? How do you know?

2. State the null and alternative hypotheses for this study.

3. Suppose that the test statistic calculated for this study yields a p value of.18. What decision should be made with regard to the null hypothesis (assume α =.05)? What can the researcher conclude about the poker players’ abilities?

4. Suppose that in reality poker players are able to predict another player’s hand with better than chance accuracy. In this case, what type of decision was made in (c) above?