Loading presentation...

Present Remotely

Send the link below via email or IM

Copy

Present to your audience

Start remote presentation

  • Invited audience members will follow you as you navigate and present
  • People invited to a presentation do not need a Prezi account
  • This link expires 10 minutes after you close the presentation
  • A maximum of 30 users can follow your presentation
  • Learn more about this feature in our knowledge base article

Do you really want to delete this prezi?

Neither you, nor the coeditors you shared it with will be able to recover it again.

DeleteCancel

Make your likes visible on Facebook?

Connect your Facebook account to Prezi and let your likes appear on your timeline.
You can change this under Settings & Account at any time.

No, thanks

The Right Questions about Statistics

This prezi gets you started in choosing the right statistics, beginning with what sort of question you want to answer and what sort of data you have to answer it with. It goes with a presentation given to University of Adelaide students.
by

David Butler

on 30 March 2015

Comments (0)

Please log in to add your comment.

Report abuse

Transcript of The Right Questions about Statistics

FROM CONCEPTS TO VARIABLES
The purpose of Statistics is to
yes or no?
what's going on?
what's the formula?
DECIDE
DESCRIBE
PREDICT/EXPLAIN
ESTIMATE
what's this number?
hypothesis tests (p-values)
descriptive statistics: graphs and basic numbers
modelling and regression
confidence intervals
How can I calculate a student's grade based on their number of hours of sleep during semester?
What sorts of things might be related to whether a person does volunteer work?
What sort of relationship might the amount of sleep a student gets have with their grades?
What possibilities are there for body temperature after a meal with or without chilli?
How many chapters do novels have?
How can I explain a person's body temperature after a meal using their temperature before and the chilli content of the meal?
How can I use a person's gender, age, income and religion to predict their chances of participating in volunteer work?
Does getting more sleep affect a students’ grades?
Are women more likely to participate in volunteer work than men?
Is your body temperature higher after a meal if it has chilli in it?
Is the median number of chapters in a novel 20?
Know the
type of question
and you can choose what
type of statistics
...
The purpose of Statistics is to
ANSWER QUESTIONS
using DATA

Know more
about your data
and you can choose what
statistical method
...
VARIABLES IN
THE DATA
HOW THE DATA
IS COLLECTED
HOW MUCH DATA
what type?
what is done to the subjects?
when is information recorded?
how are the subjects chosen?
lots of things recorded
per subject?
lots of subjects?
what distribution?
VARIABLES (things you record)
Categorical / Qualitative
Nominal
Ordinal
Numerical / Quantitative / Scale / Interval
Continuous
Discrete
(numbers: how far apart has meaning)
(words: how far apart has no meaning)
(measured)
(counted)
(ordered: "more" or "less" has meaning)
(names: "more" or "less" has
no
meaning)
Some information you need to know to figure out what stats you need.
THE RIGHT QUESTIONS ABOUT STATISTICS
Maths Learning Centre
how to measure?
DISTRIBUTION of NUMERICAL data
(how the possible values are spread out)
Skewed or worse
Approximately normal
A parametric test
will be fine
A non-parametric test might be more appropriate
KNOW
YOUR QUESTION

KNOW
YOUR DATA

HOW TO ORGANISE DATA
becomes...
missing data?
The purpose of Statistics is to
ANSWER QUESTIONS
using DATA

...
the question has to be
specific
...
the data has to be
specific

Statisticians say:
"PLEASE make it consistent!"
defining groups or measurements?
ANSWER QUESTIONS
using DATA

WHAT CATEGORIES DEFINE
Independent
Groups
Repeated
Measures
(matched pairs)
On average, how much of an effect does 30 minutes more sleep have on a students’ grades?
How much more (or less) likely is a woman to participate in volunteer work than a man?
How much higher is your body temperature after a chilli meal compared to one without?
What is the median number of chapters in a novel?
Actually it's a way to cope with VARIATION
Statistics helps to find descriptions and explanations for how things VARY, using the information from data.
This means we can use it to answer questions in situations where things do vary.
Questions are about CONCEPTS
About the relationship between concepts
About a single concept
Are science fiction books thicker than other books?
Does getting more sleep affect a students' grades?
Are women more likely to participate in volunteer work than men?
What is the median number of chapters in a novel?
How much salt is there in an avocado?
What percentage of children finish all their homework?
KNOW
YOUR STATS

This information helps you choose which stats to use.
Note: This probably doesn’t matter if you have a lot of data.
Likert scale
A non-parametric test probably won't work -- might have to treat as categorical.
OBSERVATIONAL OR EXPERIMENT
Observational study:
the only thing you did to the subjects while you were watching them was record information about them.
Experiment:
you made a choice at least once to do something that might influence the outcome (possibly you did this randomly)
RANDOMNESS
Random selection:
you chose the subjects randomly from a population, or at least they are independent of each other.
Random allocation:
you chose which subject got what treatment randomly.
OUTCOMES and EXPLANATORY VARIABLES
outcome or explanatory?
Explanatory
Variables
Outcome
Variable
(also known as
predictor variables or
independent variables)
(also known as
response variable or
dependent variable)
Hypothesis tests are designed to answer yes or no questions.
One of the answers (yes or no) is called the NULL HYPOTHESIS.
The p-value can be thought of as a measure of how consistent your data is with the null hypothesis, so a LOWER p-value is STRONGER evidence against the null hypothesis.
The SIGNIFICANCE LEVEL is the cutoff for the p-value where you decide the answer is yes or no. This is called REJECTING or RETAINING the null hypothesis.
0.01
very strong evidence
0.05
0.10
strong evidence
some evidence
no evidence
reject
retain
significance level
Confidence intervals are designed to answer "what's the number" questions.
They give a range of possible values for what a number in the population could be.
They list all of the values that mathematically seem consistent with the data.
APPEARS CONSISTENT
WITH DATA
INSIDE
NOT CONSISTENT
WITH DATA
OUTSIDE
NOT CONSISTENT
WITH DATA
OUTSIDE
HYPOTHESIS TESTS
CONFIDENCE INTERVALS
Full transcript