Statistics 212 Chapters 1 To 3

Front 1

Descriptive Statistics

Back 1

methods of organizing, summarizing, and presenting data in a convenient and informative way (graphical techniques, numerical measures, etc)

Front 2

Inferential statistics

Back 2

a body of methods used to draw conclusions or inferences of populations based on sample data. (if the population is 50,000 then take 500 [smaller number] and form an inference from the data obtained)

Front 3

Population

Back 3

the group of all items of interest to a statistics practitioner (usually large, maybe infinitely large). What it is depends (product, people, etc)

Front 4

Parameter

Back 4

descriptive measure of the population.

ex. mean number of soft drinks consumed by all students at the university; proportion who voted for Bush

Front 5

Sample

Back 5

set of data drawn from the population

Front 6

Statistic

Back 6

descriptive measure of a sample

ex. mean number of soft drinks consumed in the last week by 500 students

Front 7

Statistical Inference

Back 7

process of making an estimate, prediction, or decision about a population based on sample data.

the reliability of the conclusions we draw from that sample may not always be correct though - see confidence level and significance level

Front 8

Confidence level

Back 8

the proportion of times that an estimating procedure would be correct, if the sampling procedure were repeated a very large number of times

ex. estimate of the avg number soft drinks consumed by 50,000 student has a confidence level of 95%. aka, if we had a large number of repeated samples, estimates based of this form would be correct 95% of the time

Front 9

Significance level

Back 9

when purpose of the statistical inference is to draw a conclusions about a population, this measurement determines how frequently the conclusion will be wrong in the long run

Front 10

variable

Back 10

some characteristic of a population or sample

ex. mark on a stat exam is a char of stat exams interesting to stat students (because not everyone gets the same score as it varies)

Front 11

values

Back 11

possible observations of the variable

ex. exam mark betw. 0 to 100

Front 12

data (datum - for one)

Back 12

the observed values of a variable

ex. midterm test scores of 10 students

Front 13

interval data

Back 13

real numbers, such as heights, weights, incomes, and distance. also refer to these type of data as quantitative or numerical

graphically, usually use a histogram

Front 14

nominal data

Back 14

these data have values put in categories. also known as qualitative or categorical data

ex. marital status
graphically, usually uses bar graph and pie chart or even a scatter diagram

Front 15

ordinal data

Back 15

appear nominal, but values are in order

ex. evaluating a course as good, fair, poor, etc.

Front 16

hierarchy of data

Back 16

interval may be treated as ordinal or nominal, but these lower-level data types may never be treated as what is higher; aka deals with convertibility

ex. marks in a course (interval) are same as letter grades (ordinal) which is same as pass/fail (nominal)

Front 17

frequency distribution

Back 17

summarization of data in a table that presents the categories and their counts

ex. bar chart (nominal), histogram (interval)

Front 18

relative frequency distribution

Back 18

lists the categories and the proportions with which each occurs

ex. pie chart (nominal), ogive (interval)

Front 19

classes

Back 19

series of intervals that cover the complete range of observations (are counted when creating a frequency distribution)

Front 20

number of class intervals

Back 20

either depends on a table given, or it is equal to 1 + 3.3log(n) where n is the number of obervations

Front 21

class width

Back 21

(largest observation - smallest observation) / (number of classes)

Front 22

symmetry

Back 22

if we draw a vertical line down the center of a histogram and the two sides are identical in shape and size there is said to be ______

Front 23

positively skewed

Back 23

a skewed histogram with the long tail extending to the right

ex. many low paid employees and few high

Front 24

negatively skewed

Back 24

a skewed histogram with the long tail extending to the left

ex. time it takes to finish tests (few finish quickly)

Front 25

modal class

Back 25

the class with the largest number of observations

Front 26

unimodal/bimodal histogram

Back 26

a histogram with one/two peak(s); if two, they're not necessarily same in height

Front 27

stem-and-lead display

Back 27

used to offset the drawbacks of a histogram which does not allow you to see most of the observations. it has two parts (are part of name so can't tell)

Front 28

ogive

Back 28

a graphical representation of the cumulative relative frequencies (which highlights the proportion of observations tat lie below each of the class limits)

Front 29

univariate v. bivariate

Back 29

graphical/tabular techniques to summarize single sets of data (termed ____) v. those applied to depict relationship between variables (termed ____)

Front 30

contingency table (cross-classification or cross-tabulation table too)

Back 30

used to describe the relationship between two nominal variables; can use these to then graph bar chart

ex. frequency of what newspaper read and what occupation reader has

no relationship means patterns in bar charts approximately same; differences indicate relatioinship

Front 31

scatter diagram

Back 31

used to describe the relationship between two interval variables

Front 32

independent v. dependent variable (scatter diagram)

Back 32

where one variable depends to some degree on the other variable, we label Y the dependent and X the independent

Front 33

linear relationship (strong, medium, or weak)

Back 33

to determine strength of dependency of the variables we draw a straight line through the points (later on we use the least squares method)

If most of the points fall close to the line then there is said to be a _______

Front 34

positive linear relationship

Back 34

when one variable increases, so does the other

Front 35

negative linear relationship

Back 35

when two variables tend to move in opposite directions

Front 36

no relationship; nonlinear relationship

Back 36

when there is no direction among two variables; also when the relationship is not linear

Front 37

cross-sectional data

Back 37

classifying data according to whether the osbservations are measured at the same time

ex. real estate consultant selling houses and shows the specific form of the function by showing the 100 most recently sold with their price, size, age, and lot size

Front 38

time-series data

Back 38

classifying data according to whether they represent measurements at successive points in time

Ex. real estate consultant works to forecast the monthly housing starts and collects it for the region in the past 5 years

Front 39

line chart

Back 39

used to depict time-series data, plotting a variable over time (time is on the horizontal axis)

Front 40

graphical excellence (look at hint to see more)

Back 40

term applied to techniques that are informative and concise and that impart information clearly to their viewers.

Hint 40

1. graph presents large data sets concisely

2. ideas and concepts statistics practitioner wants to deliver are clearly understood by the viewer

3. graph encourages viewer to compare two or more variables

4. display induces viewer to address substance of data and not form of graph

5. no distortion of what data reveal

Front 41

Graphical deception

Back 41

Several examples:
1)Graph without vertical scale
2)different captions for same graph
3)Perspective not correctly applied
4)Zoom in/out of vertical or horizontal axis
5)Width of bars (also, correctly using pictograms)

Front 42

Written report

Back 42

1) State purpose of statistical analysis & include possible results of experiment/decisions that might follow

2) Describe experiment - know reader & assure them about a properly conducted experiment

3) Describe your results - use charts & tables, have clarity/brevity, precise, honest, etc.

4)Discuss limitations of statistical techniques, because rarely is it definitive

Front 43

Oral Presentation

Back 43

1) Know audience
2)Restrict you main points to objectives, conclusions and recommendations of the study
3)Stay within your time limit
4)Use graphs (tables)
5)Provide handouts (makes things easier for audience to follow)