Creating User Interfaces

1
Creating User Interfaces

• Review midtermSampling
• Homework User observation reports due next week

2
Sampling

• Basic technique when it is impossible or too
  expensive to measure everything/everybody
• Premise possible to get random sample, meaning
  every member of whole population equally likely
  to be in sample
• NOTE not a substitute for monitoring directly
  activity on / with interface

3
Source

• The Cartoon guide to Statistics by Larry Gonick
  and Woollcott SmithHarperResource
• Procedures (formulas) presented without proof,
  though, hopefully, motivated

4
Task

• Want to know the percentage (proportion) of some
  large group
• adults in USA
• television viewers
• web users
• For a particular thing
• think the president is doing a good job
• watched specific program
• viewed specific commercial
• visited specific website

5
Strategy Sampling

• Ask a small group
• phone
• solicitation at a mall
• Follow-up or prelude to access to webpage
• other?
• Monitor actions of a small group, group defined
  for this purpose
• Monitor actions of a panel chosen ahead of time
• ALL THESE make assumption that those in group
  are similar to the whole population.

6
Two approaches

• Estimating with confidence intervalc in general
  population based on proportionphatin sample
• Hypothesis testingH0 (null hypothesis) p p0
  versusHa p p0

7
Estimation process

• Construct a sample of size n and determine phat
• Ask who they are voting for (for now, let this
  be binomial choice)
• Use this as estimate for actual proportion p.
• but the estimate has a margin of error. This
  means The actual value is within a range
  centered at phat UNLESS the sample was really
  strange.
• The confidence value specifies what the chances
  are of the sample being that strange.

8
Statement

• I'm 95 sure that the actual proportion is in the
  following range.
• phat m
• Notice if you want to claim more confidence, you
  need to make the margin bigger.

9
Image from Cartoon book

• You are standing behind a target.
• An arrow is shot at the target, at a specific
  point in the target. The arrow comes through to
  your side.
• You draw a circle (more complex than/- error)
  and sayChances arethe target point is inthis
  circle unless shooterwas 'way off' . Shooter
  would only be way off X percent of the
  time.(Typically X is 5 or 1.)

10
Mathematical basis

• Samples are themselves normally distributed
• if sample and p satisfy certain conditions.
• Most samples produce phat values that are close
  to the p value of the whole population.
• Only a small number of samples produce values
  that are way off.
• Think of outliers of normal distribution

11
Actual (mathematical) process
Sample size must be this big

• Can use these techniques when np5 and
  n(1-p)5
• The phat values are distributed close to normal
  distribution with standard deviation sd(p)
• Can estimate this using phat in place of p in
  formula!
• Choose the level of confidence you want (again,
  typically 5 or 1). For 5 (95 confident),
  look up (or learn by heart the value 1.96 this
  is the amount of standard deviations such that
  95 of values fall in this area. So .95 is
  P(-1.96

12
Notes

• p is less than 1 so (1-p) is positive.
• Margin of error decreases as p varies from .5 in
  either direction. (Check using excel).
• if sample produces a very high (close to 1) or
  very low value (close to 0), p (1-p) gets
  smaller
• (.9)(.1) .09 (.8)(.2) .16, (.6)(.4) .24
  (.5).5).25

13
Notes

• Need to quadruple the n to halve the margin of
  error.

14
Formula

• Use a value called the z transform
• 95 confidence, the value is 1.96

15
(No Transcript)
16
Mechanics

• Process is
• Gather data (get phat and n)
• choose confidence level
• Using table, calculate margin of error.
• Book example 55 (.55 of sample of 1000) said
  they backed the politician)
• sd(phat) square_root ((.55)(.45)/1000)
  .0157
• Multiply by z-score (e.g., 1.96 for a 95
  confidence) to get margin of error
• So p is within the range .550 (1.96)(.0157)
  and .550 (1.96)(.0157)
• .519 to .581 or 51.9 to 58.1

17
Example, continued

• 51.9 to 58.1
• may round to 52 to 58
• or
• may say 55 plus or minus 3 percent.
• What is typically left out is that there is a
  1/20 chance that the actual value is NOT in this
  range.

18
95 confident means

• 95/100 probability that this is true
• 5/100 chance that this is not true
• 5/100 is the same as 1/20 so,
• There is only a 1/20 chance that this is not
  true.
• Only 1/20 truly random samples would give an
  answer that deviated more from the real
• ASSUMING NO INTRINSIC QUALITY PROBLEMS
• ASSUMING IT IS RANDOMLY CHOSEN

19
99 confidence means

• Give fraction positive
• Give fraction negative

20
Why

• Confidence intervals given mainly for 95 and
  99??
• History, tradition, doing others required more
  computing.

21
Let's ask a question

• How many of you watched the last Super Bowl?
• Sample is whole class
• How many registered to vote?
• Sample size is number in class 18 and older
• ????

22
Excel columns A B
23
Variation of book problem
Divisor smaller

• Say sample was 300 (not 1000).
• sd(phat) square_root ((.55)(.45)/300)
  .0287
• Bigger number. The circle around the arrow is
  larger. The margin is larger because it was
  based on a smaller sample. Multiplying by 1.96
  get .056, subtracting and adding from the .55 get
• .494 to .606You/we are 95 sure that true
  value is in this range.
• Oops may be better, but may be worse. The fact
  that the lower end is below .5 is significant for
  an election!

24
Exercise

• size of sample is n
• proportion in sample is phat
• confidence level produces factor called the
  z-score
• Can be anything but common values are 80, 90,
  95, 99)
• Use table. For example, 95 value is 1.96 99
  is 2.58
• Calculate margin of error m
• m zscore sqrt((phat)(1-phat)/n)
• Actual value is phat m and

25
Opportunity sample

• Common situation
• people assigned/asked to have a meter attached to
  their TVs
• people asked/voluntarily sign up to have a meter
  (software) installed in their computers.
• people asked during a Web session to participate
  in survey
• students in a specific class!
• Practice is to determine categories
  (demographics) and project the sample results to
  the subpopulation to the population
• For example, if actual population was 52 female
  and 48 male, and sample (panel) is 60 male and
  40 female, use proportions to adjust result
• But maybe this fact hides problem with the sample
• Has negative features of any opportunity sample
• Are these folks different than others in their
  (sub)population?

26
Requirements

• Model / Categories must be well-defined and valid
• Hispanic versus (Cuban, others) in Florida in
  2000
• Need independent analysis of subpopulations
  representation in general population
• The sample sizes are the individual Ns, making
  the margin of errors larger

27
Adjustment from panel data

• Panel of 10 6 females, 4 males
• Population is 52 female and 48 male
• Female panelists 5 liked interface, 1 didn't.
  Male panelists 2 liked interface, 2 didn't.
• Estimate for whole population (size P)
• (5/6) .52 P (2/4).48 P

28
Critical part of surveys

• and survey analysis
• Understand the exact wording of question.
• Understand definition of categories of
  population.
• Don't make assumptions
• Admire Michelle Obama example
• Belief in Holocaust example

29
Usability research

• Often aims for qualitative, not quantitative
  results
• Ideas, critical factors
• Note there are fields of study
• Non-numeric statistics
• Qualitative research
• Still necessary to be systematic.
• AD consider taking Statistics!

30
Homework

• Continue work on user observation studies
• This is qualitative work