# SOC222H5 Lecture Notes - Lecture 6: Confidence Interval, Null Hypothesis, Point Estimation

## Document Summary

Applying the concept of standard error for estimation. Estimation: something that is available in a sample. Confidence intervals are related to the empirical rule. From 3000 sample spend 4 hrs volunteering, larger sample better results. When we want to make an estimate we need to know how confident you are about the value lieing. If the standard deviation is not given use the sample information. When the sample size decreases the sample error increases. Collective evidence and weighing evidence, it is not about finding the truth, we reject or fail to reject-do we have enough statistical evidence. Rejecting the null is saying that this is statistically significant. Negative makes the distribution to the left side. If it says greater than or less than it is a one tail. Having not enough evidence, and want to make sure that they earn different, there is a probability of getting from either side therefore two tailed test.