Math 9 Data Analysis Project by Devon Nicholson and Sydney Cromarty. Click the link to see it in Word: Data Analysis Project
Biasedness may influence the data if a survey is showing preference to one of the choices than the others.
Ex. The majority of people in Canada believe that all drugs should be illegal. Do you agree?
Ex. Even though smoking causes health issues, and is terrible for the body, do you like smoking?
The use of language may influence the data if the question is written in a way that customers will misunderstand, think differently of, or be ushered towards a certain answer.
Ex. Don’t you think that Bobby Oar is the best hockey player ever?
Ex. The majority of people don’t think coffee doesn’t have unhealthy chemical.
Ethics may influence the data by having inappropriate behaviours, leave out certain data or statistics to prove a point.
Ex. A company electrocutes people when they do not answer the question “correctly”.
Ex. 90% of Americans eat at least once a week (the true question was a survey on which fast food chains you go to, not just McDonald’s).
The cost may influence the data because it may cost more to conduct the survey itself than the benefits that come with it.
Ex. A free range beef company is setting up a table to promote their beef over normal beef. They must pay for a banner, a table, space, rights to publicize, staff to work, beef, etc.
Ex. A&W sets up a booth at the mall to taste test their newest burger. They must pay for advertisement, space, works, burgers, etc.
Time and timing may affect the data because it could be collected at time that would beneficial to asker of the question.
Ex. A survey was conducted after the Haiti earthquake; Haitians believe they need more protection and supplies from other countries including the USA to prevent other disasters. A survey asks pregnant women how well Trojan condoms work.
Privacy may affect the data because some customers may be forced into answering or their information may not be kept confidential.
Ex. Ikea surveys people about where their newest store should be, they ask the surveyors for their address. After the survey is done they share that private address with those that may be interested.
Ex. Costco is doing a survey about their taste testing booths, they ask the surveyors for their phone numbers as proof they aren’t robots. Costco then sends those numbers to spam callers or whoever is interested in contacting people.
Culture sensitivity may affect the data because the question could be offensive to some cultures.
Ex. Superstore goes to a Muslim church during Ramadan and ask them what their favourite food is from a list of spaghetti, pizza, or tacos.
Ex. A math teacher asks all the Asian students their favourite math unit from the year.
Population is when everyone from a group of people is surveyed.
Ex. All students of a high school are asked their opinion on the student council.
Ex. An entire U12 soccer team was surveyed about how their coach performed this year.
Sample is when some of the people from a group are chosen to be surveyed.
Ex. All the grade 3s from an elementary school were asked about their favourite recess sport.
Ex. All the people who watch Doctor Who on BBC were asked doctor they liked the most.
Convenience Sample is when a survey is conducted on people who are easily accessed. This is beneficial because it is an easy and quick way to get customers input.
Ex. After a marathon, the hosts interview the first 10 finishers to ask about what they thought of the race.
Ex. Subway asked the first 20 people buying a 6-inch turkey bacon sandwich about what they thought of it.
Random sample is when you randomly pick a set number of people from a group to answer your question. This is beneficial because you have no knowledge of who could be surveyed, opinions could vary widely which could give more feedback.
Ex. 7 Pittsburgh Penguins fans were randomly chosen through a generator to be asked what they thought of the team so far.
Ex. 5 preschool students were drawn from a hat to be surveyed about their favourite toys at school.
Stratified Sample is when the population is split into groups and only some of the members of the group are chosen to be surveyed. This is beneficial because it is a more reasonable amount of people to be surveyed that could give you an honest opinion.
Ex. High school students are divided into groups depending on their grade, from there a maximum of 20 students can vote for what cafeteria food will be served on Thursdays.
Ex. A senior center homes divides all of their elderly into groups depending on their birth date, from there a maximum of 3 are chosen to vote on what should be added to the home to make it a safer a friendlier environment.
Systematic Sample is choosing people in intervals or patterns from a list. This is beneficial because you can’t cheat to get certain people, you get whoever your interval landed on. This will let you get a diversity of answers from random people.
Ex. Every 3rd student will be asked about cafeteria food when leaving.
Ex. People in every 2nd house down Mary Hill street will be surveyed on what they think of their neighbourhood.
Voluntary Sample is when a survey is conducted that allows anyone to participate. This is beneficial because it allows everyone from the public to answer which gives the questioners a reliable answer if the majority of their customers voluntarily answer.
Ex. Facebook sets up a booth for customers to be surveyed on their overall experience. Anyone can come and participate as long as they have an account.
Ex. Tim Horton holds a survey for all of their customers on what their favourite doughnut is.
Theoretical Probability is what should theoretically should happen while doing an activity.
Ex. You role a die six times, it lands on six 1/6 of the times.
Ex. You flip a two-sided coin 4 times; it lands on tails 2/4 of the times.
Experimental Probability is the ratio of what happens from the number of trials or times you perform the activity.
Ex. You role a die six times, you land on six 3/6 of the times.
Ex. You flip a two-sided coin 4 times; it lands on tails 1/4 of the times.
This chart is supposed to be a job loss graph but instead the data is from the total number of unemployment in America at that time. This is misleading because it is not answering the question. The number of unemployed Americans has nothing to do with the job loses from December 2007 to June 2010. The graph also doesn’t start at 0.
This chart is misleading in many ways. Firstly, he percentage does not equal 100% instead it is at 120. They also removed and mashed together some of the data that didn’t fit into their criteria. Originally there were 5 groups (Very likely, somewhat likely, not very likely, not very likely at all, and an unknown section). Fox news added the original “somewhat likely” and “very likely” to get 59%. They then put very likely in as it’s own group even after mixing it together with another. “Not very likely at all” and “not very likely” were mashed together to get 26% and all those unsure were removed.
This graph is misleading because Fox News has exaggerated the prices of taxes. They counted the taxes more than once, by adding the tax price on top of the regular price when in reality it was already included.