This unit is a formal introduction to statistical inference where you will see building blocks from the previous units come together in commonly used statistical inference methods just as confidence, intervals and hypothesis tests. In this until will also introduce the central limit theorem which provides the basis for these methods. Let's start with an example for a survey conducted by the Pew Research Center. The study is titled Young, Underemployed, and Optimistic, Coming of Age, Slowly, in a Tough Economy. Young adults hit hard by the recession. A plurality of the public. 41% believes young adults rather than middle aged or older adults are having the toughest time in today's economy. Tough economic times, altering young adults daily lives and long term plans. While, negative trends in the labor market have been felt most acutely by the youngest workers. Many adults in their late 20s and early 30s have also felt the impact of the weak economy. Among all 18 to 34 year olds, fully half, 49%, say they have taken a job they didn't want just to pay the bills with 24% saying they have taken an unpaid job to gain work experience. We're also told that the general public survey is based on telephone interviews conducted between December 6 and 19 2011 with a nationally representative sample of 2,048 adults aged 18 and older living in the continental United States. Margin of sampling error is plus or minus 2.9% points for results based on the total sample and 4.4% points for adults ages 18 to 34 at the 95% confidence level. With all sorts of studies the margin of error is often recorded and this information can often be found in the fine print. So, what does this mean? Remember, the study had estimated that 41% of the public believes that young adults, rather than middle aged or older adults, are having the toughest time in today's economy. Meaning that, 41 plus or minus 2.9. We are 95% confident that 38.1% to 43.9% of the public believe young adults rather than middle-aged or older adults are having the toughest time in today's economy. We were also told that 49% of the public had taken a job they didn't want just to pay the bills. Meaning that 49 plus or minus, 4.4%, we are 95% confident that 44.6 to 53.4% of 18 to 34 year olds have taken a job they didn't want, just to pay the bills. The 41 and 49% we have on hand, come from the sample data. But we're often interested in population parameters. Since compete populations are difficult, or impossible to collect data on. That would mean collecting data from the entire US population for this study for example, we use sample statistics as point estimates for the unknown population parameters of interest. But these samples statistics vary from sample to sample. Other random samples of Americans would yield slightly different estimates, for example. Quantifying how sample statistics vary provides a way to estimate the margin of error associated with our point estimate. Discussion on sampling variability, in other words, how estimates vary from one sample to another, will be the first focus of this unit. Within that discussion, we're going to introduce the central limit theorem which describes shapes, centers and spreads of sampling distributions when certain conditions about the population, as well as the sampling scheme are met. This unit is also going to serve as a formal introduction to statistical inference. And we're going to highlight techniques that are considered to be pillars of statistical inference, such as confidence intervals and hypothesis tests. We will wrap up the unit with a discussion of statistical versus practical significance. We're going to consider the effect of sample size, confidence, and significance levels in our conclusions. And finally we're going to discuss statistical power.