Observations that you or someone else records Data is more than numbers; it is numbers in context Data isn’t only numbers; it is also the story.

Math 140: Chapter 1, Introduction to

Data

Discuss within your group what we can say about this data; 1 minute

11, 8 ½ , -13, 174.3, , , 0

Share out…

Here’s some data…

Observations that you or someone else records

Data is more than numbers; it is numbers in context

Data isn’t only numbers; it is also the story behind the numbers

Data is/are…

Talk for 30 seconds and then share out

News media; surveys galore!

Any recent news report (based on survey or poll) that you can recall? Radio? TV? Internet?

Recent survey showed that of the people who use Twitter and Facebook, more are likely to get their news from Twitter than Facebook

More than half of Jewish students at American colleges reported personally experiencing or witnessing anti-Semitism within the past six months, according to survey findings.

Who collects data?

Variables record characteristics of people or things

# people in the survey who said they get their news from Facebook primarily (or Twitter primarily)

# of Jewish college students who were (or were not) observing or experiencing anti-Semitism

Characteristics, variables …

Sample Statistic(changes from

sample to sample; can vary)

Examples include

Population Parameter(is fixed; does not

change)

Examples include µ, p,

σ

Sample vs. Population …

Always, always comment, answer, compare, contrast… whatever the case.. in context

What are the objects? What was measured? What are the units of measure?

Example… Researchers studied the amount of tofu, in pounds, that a typical American consumes per year.

Context is key!

Consumer beware…

Questions to ask yourself before you believe a poll/survey... Who carried out the survey?

How was the sample selected?

How large was the sample?

What was the response rate?

How were the subjects contacted?

When was the survey conducted?

What was the exact question asked?

Come up to board and write where you last went out to eat and how much you spent there (approximately).

What are our variables?

Two types of variables: Numerical (quantitative) and categorical (qualitative)

Which is which for our data?

Two types of variables…

Can you think of a categorical data that looks like numerical data… but it isn’t. It’s really categorical. Discuss with your group for a minute…

Zip codes, social security numbers, phone numbers, student ID numbers, etc.

Always ask yourself, does finding the mean (average) of this data make sense?

Sometimes it’s in camo…

Come up to the board and write yes if you have pierced ears and no if you don’t. Use a blue marker if you are male; use a black marker if you are female.

What trends do you see?

Hard to clearly see trends when data is not organized

One common way to organize categorical data is in a 2-way table

Categorical data…

Ears pierced?

Ears Pierced Ears NOT Pierced

Male

Female

Suppose on a given day we randomly ask 200 Disneyland visitors if they have been on Space Mountain yet that day.

23% said they had been on Space Mountain already. How many of these visitors had been on Space Mountain already?

46 of these Disneyland visitors had already been on Space Mountain.

Finding ‘how many’ from percents…

Suppose we randomly ask another group of Disneyland visitors if they have been on Thunder Mountain yet that day. 14% tell us yes, and this is equal to 42 visitors.

How many total visitors were in our survey?

300 visitors were asked if they had been on Thunder Mountain yet that day.

Finding ‘how many’ from percents…

Sport Injuries Participants

Baseball 178,668 15,600,000

Basketball 615,546 28,900,000

Bowling 21,133 43,900,000

Football 387,948 17,700,000

Ice Hockey 16,435 2,100,000

Soccer 178,519 14,500,000

Softball 125,875 13,600,000

Tennis 19,633 11,000,000

Volleyball 59,225 11,500,000

Rates vs. Raw Numbers…

There are always firefighters at fires. Therefore, firefighters cause fires

x: Person regularly attends religious services y: How long a person lives

x: Number of ministers y: Rum imports

x: HS seniors’ SAT scores y: Students’ first year GPA

Causality vs. Association

x causes y, there must be a controlled, randomized, well-designed experiment

Treatment (explanatory, factor) variable; Response (outcome) variable

Sample size ‘large’

Treatment group (receiving treatment(s); control group (receiving placebo)

Random assignment of subjects (or experimental units)… using an acceptable randomization process

Double-blinding is best

To even consider using the phrase…

Reduce bias

Bias: Systemically ‘off’

Examples: scale, clock

Goal:

Gastric freezing is a proposed treatment for ulcer pain in the upper intestine. In this treatment the patient swallows a deflated balloon with tubes attached. Then a cold liquid is pumped through the balloon for an hour.

The rationale is that the cooling will reduce production of acid and relieve ulcer pain. A study reported in the Journal of the American Medical Association provided gastric freezing to a large number of patients and reported that gastric freezing reduced acid production and relieved stomach pain.

Based on this the (safe and easy) treatment was used for several years.

A later study divided ulcer patients randomly into two groups. The first group was treated by gastric freezing. The second group received a placebo treatment in which the liquid was at body temperature rather than cooled

In the first group 34% of 82 patients improved. In the second group 38% of 78 patients improved. Based on this second study gastric freezing was discontinued as a therapy for ulcers.

Gastro follow-up…

Is it well designed? Why or why not?

Marshmallow videohttps://www.google.com/?gws_rd=ssl#q=jimmy+kimmel+marshmallow+test

Why is this an experiment?

https://www.youtube.com/watch?v=RK-oQfFToVg

Jimmy Kimmel… “experiment”

Is this an observational study or a controlled experiment?

Experiment? Was there a large sample size? Was randomization used to assign participants to treatment groups? Was the study double-blinded? Was there a placebo?

Was the paper published in a peer-reviewed journal or just posted on internet?

Did the study follow the subjects for a long period of time?

Again, consumer beware…

We will randomly assign you into partners

Minitab – calc – random – integer

Move to sit next to your partner

Exit Ticket…

You and your partner create a list of a total of 3 questions we may use for a survey (data we may collect) this semester; be specific; be “G” rated

Next to each question, write the type of data your question is asking for (categorical or numerical) and justify why it is that type of data

Example: What do you weigh in pounds? This is quantitative data.

But both your names on the paper; write neatly

Exit Ticket…