A convenience sample is when people are chosen based on the fact that they're easy to talk to. However, it usually creates underrepresented data. If I went to a football field and asked everyone what they're favorite sport is, it would likely be an unrepresentative sample because they all probably love football.

Wording Bias is when a question is worded specifically to make people answer the question in a certain way. If I asked people "Do you believe in aliens?" most people would say no. However, if I asked "Do you believe in the potential for intelligent life somewhere in the universe?" the responses would be quite different.

Bias is the systematic changing of parameters in a population in order to effect the samples outcomes in a favorable way. For example if you wanted to see how many people enjoy meat and you sample at a vegetarian resturant, there will be a bias because the only people in the survey do not like meat.

A Simple Random Sample, or SRS, is when any individual in a population has an equal chance of being chosen, and every set of individuals from the sample has an equal chance of being chosen. If we wanted to take a a random 3 kids from this class, we could write everyone's names down on a piece of paper and choose 3, giving us a SRS. An SRS is a truly random sample because everyone has an equal chance of being chosen. However, it is difficult to use an SRS with a very large group of people.

A Stratified Simple Random Sample is when a large population is separated into more similar individuals into "strata." From these strata, we can take an SRS from each strata to get a representation of the entire population. A benefit to strata is that specific groups can be more accurately represented, and that all different groups can be represented equally by being split up. However, clusters may have less variability because they are from specific groups.

A Cluster Sample is when a population is divided into smaller groups that are meant to represent the total population. Although it is easier to take a cluster sample because we will work with smaller groups of people, it is difficult to find a cluster that accurately represents the total population.

A voluntary response sample is when people make the decision to respond to a general question to the public. These samples usually are biased because people who have a strong opinion on the question are more likely to respond than people who don't really have an opinion. For example, people who respond to political phone surveys are not necessarily representative of the population because they feel strongly about political issues.

Under-coverage occurs when a sample is unrepresentative of a population. This can occur through many ways of sampling errors. For example, if I went to a student council meeting and asked everyone how they feel about participating in school activities, most would say they loved to participated. However, this wouldn't accurately represent the total population of Duxbury High School because most people may not do any activity.

Bias always has a direction, or inclination in favor of one thing or another. Bias also always has a source, or the reason that a sample has bias and the reason that it favors a certain outcome. An example of this is when you have a survey to find out the number of high school students who own a football helmet, and you only ask members of the football team. Bias is evident here and the direction of the bias is in favor of the number of people who have a football helmet. The source of this bias is that you only asked members of the football team, who all obviously own a football helmet.

Response Bias happens when people deliberately lie during surveys. For example, if people are asked if they have been pulled over before, people may answer "no" because they want to seem like better drivers.

A nonresponse occurs when certain individuals can't be reached or refuse to respond to a survey. If a company is doing a phone survey and only call during 11-2 every day, people who work away from their house won't be reached.

Should the DHS prom venue be at Gillette Stadium or at the New England Aquarium?
For our survey, we will ask the senior class at Duxbury High School which venue they prefer using a simple random survey. There are 250 students in the senior class, and each student will be given a number from 001-250 alphabetically by last name. Next, we will randomly choose a line to start at on the random number table. After this, we will read through each line and search for 3 digits at a time. Every number between 001 and 250 would represent a digit of success, but every number from 251 to 999 is a failure. There cannot be any duplicate numbers, so a repeated number would be a failure. Each digit of success will be recorded, and after we reach 50 students from the sample we will survey them and ask which venue they prefer.