9.57 Email, stress, and a paired-samples t test: Researchers wondered if frequent checking of email increases stress (Kuskley & Dunn, 2015). They randomly assigned half of their participants to check email just three times a day for a week, and then in the second week, to check email as often as they wanted. The other half of participants checked their email as much as they wanted in the first week, and just three times a day in the second week. The researchers did find that participants were less stressed, on average, during the limited email week than during unlimited email week. But they wanted to be sure that it really was limited email that led to this effect. So, they conducted a paired-sample t test to be sure that participants were doing as they were told. The researchers reported that "confirming the success of our manipulation, people checked their email significantly fewer times per day in the limited email condition (M= 4.70, SD= 4.10) than in the unlimited email condition (M= 12.54, SD= 8.02; t (115) = - 10.23, p < .001). "Was this because those who checked less frequently ended up getting less email or just ignored a lot of potentially stressful email? It doesn't seem like it. The researchers also reported that "there were no significant differences between conditions in how many emails people received (M_limited = 16.64 vs. M_unlimited = 16.04, t(114) = 1.31, p =.19) or responded to (M_limited= 5.30 vs. M_unlimited = 5.95, t(115) = -1.58, p = .12), suggesting that our manipulation primarily affected how often people checked email rather than the volume of email they managed."

1. Why did the researchers use paired-samples t tests to explore their concerns about exactly what might be affecting stress levels in their study?
2. Explain why it would have been useful for the researchers to report confidence intervals in addition to the results of hypothesis testing.
3. Explain why it would have been useful for the researchers to report effect size in addition to the results of hypothesis testing.
4. Explain how the researchers' additional hypothesis tests, as described here, help them rule out some potential confounds.