91Ó°ÊÓ

Understanding the difference between a population and a sample is fundamental in statistics and has a direct impact on research efficacy. A population encompasses every member of a specified group whereas a sample is a subset of that population.

For instance, in the aforementioned Gallup poll scenario, the population includes all adults living in the United States. The sample, on the other hand, consists of the 1,021 adults actually surveyed by Gallup in 2017. Selecting a representative sample is key, as it's impractical to survey entire populations. This is where sampling techniques come in, with randomized selection ensuring that each member of a population has an equal chance of being included, thus helping to mitigate bias.

The term parameter of interest refers to a summary number that describes a feature of the entire population. It's a value that we aim to estimate or learn more about, through the data collected from our sample.

In the context of the Gallup poll, the parameter of interest is the true proportion of all adults in the U.S. who support a ban on smoking in public places. This number is relevant to policymakers and public health entities shaping regulations. However, we cannot feasibly obtain this information from the entire population, which is why researchers rely on estimates calculated from sample data, hoping they closely mirror the true population parameter.

The field of inferential statistics involves making predictions or inferences about a population based on the information gleaned from a sample. It helps us to estimate population parameters, test hypotheses, and make decisions.

Using the principle of inferential statistics, the Gallup poll's finding that 57% of the sampled adults support a smoking ban becomes a cornerstone for hypothesis testing. Statistics obtained from the sample—like the 57% statistic in the exercise—serve as estimators for the true population parameters. Inferential statistics encompasses various techniques, such as confidence intervals and hypothesis tests, to measure the reliability of our estimates and ensure that the conclusions drawn can be generalized to the whole population with a transparent level of uncertainty.