Ryan Roths

ITMC

Response paper

March 12, 2009

In chapter 4 of Asher’s book, he talks about several different types of sampling techniques. He says, “Sampling is the selection of a subset of respondents from a broader population.” If you conduct the sampling correctly, then you should be able to represent the population to the best that you can. According to Asher, probability sampling is what makes a survey scientific. By using probability sampling, you then will be able to find out the sampling error and the confidence intervals. Sampling error and confidence intervals will help give the readers more peace of mind when looking at the surveys. If you get a large sampling error or your confidence interval is not reliable, then you will not be able to represent the true population the best that you can. You can make sure to represent the population by picking a sufficient sample size, and by random sampling. The response rate of your survey is also very important when trying to represent the population.

Asher then goes over the sampling designs that are available. The first one he describes is simple random and systematic sampling. Simple random sampling means that everyone in the population has an equal chance of being selected to participate in the survey. By using this type of sampling it ensures the probability that your survey is representative to the population of interest. The next style of sampling Asher discusses is stratified sampling. This is when the population is divided into smaller groups, or strata according to the interests of the pollster. The strata guarantee is that there will be a significant amount of at least one specific group sampled to produce more accurate data about that sample. The last kind of sampling Asher discusses is cluster and multistage sampling. This is when samples are taken according to your geographical location. This is cost efficient to the pollster and convenient at the same time.

One of the major sampling designs that does not get enough attention is phone sampling. There are several advantages when conducting a phone sampling interview. The first advantage is the speed at which it can be conducted. According to Asher, “personal interviews and mailed questionnaires take too much time.” Another advantage is that a phone sampling interview is relatively cheaper to conduct than other types of questionnaires or interviews. The last advantage that Asher discusses is that phone interviews are less personal or invasive to conduct than a one on one interview. I would be more likely to conduct a phone interview than going somewhere to conduct a one on one interview.

After the discussion on phone sampling techniques, he discusses sampling error. Asher says that most people are skeptical on how a very small sample size relative to the whole U.S. population, can predict the attitudes that represent the whole population. To explain this he goes over statistical theory and probability sampling again. He uses an analogy to explain how this relationship works. With the example, to perform a blood test, a medical technician only needs a drop or two of blood. This is a small sample size of that person’s blood, but since one drop of blood has the same properties as all the blood in your body, the drop is as representative of the true population as you are going to get.

on March 14, 2009 at 5:22 am |iowajournalism“your confidence interval is not reliable”—-what does this mean?