Data Analysis = Convert raw data into useful information to gain insight and make decisions.

Synonyms: Data Analytics, Analytics, Business Intelligence, Statistical Analysis, Mathematical Analysis, Data Analysis
Goal: help make data-driven decisions, which tends to be more accurate and help to achieve goals more consistently

Raw Data = data stored in its smallest form in a cell

Not Raw Data:
Data about Google Stock
10/25/2021, $2,775.46, 1,054,085, Google
Raw Data:
Date Close Volume Company
10/25/2021 $ 2,775.46 1,054,085 Google

Proper Data Set = Data Set = Table

Table is made up of: Textbook definitions:
Field = column in table Data = Facts and figures collected, measured, summarized, and analyzed for presentation and interpretation.
Field Name= name at top of field that describes what data goes into field Table = field names in first row, records in subsequent rows
Record = one row in table Elements = unique list of entities on which variable data are collected
Each company name is an element in this element field
Table requirements in Excel:
1) Field names in first row Variable = a characteristic or attribute of interest for the element
2) Records of related data in subsequent rows There are 5 variable fields in this table
3) Empty cells or Excel Row/Column Headers, all the way around table

Close Price Volume Company Fields Company Ticker Year Incorporated Industry No. Employees P/E Ratio ($)
$2,775 1,054,085 Google Records Google GOOG 2015 Software 150,028 28.12
$2,793 1,412,937 Google Amazon AMZN 1996 Retail 1,335,000 58.72
$2,929 2,592,546 Google Microsoft MSFT 1993 Software 181,000 36.84
$2,923 1,620,903 Google . Ford F 1919 Auto 186,000 20.09
$2,965 1,447,725 Google . Caterpillar CAT 1986 Machinery 97,300 21.82
$3,320 2,225,956 Amazon .
$3,376 2,698,342 Amazon There are 6 fields in the above table: 1 element field and 5 variable fields
$3,392 2,702,224 Amazon For each record there are 6 cells: 1 element cell and 5 variable cells
$3,447 5,708,733 Amazon
$3,372 6,486,077 Amazon Observation = variable data for one element
$308 17,554,469 Microsoft Observation for the element Google = GOOG 2015 Software 150,028 28.12
$310 28,107,349 Microsoft
$323 52,588,690 Microsoft Measurements are made for each variable to provide data in the observation.
$324 26,297,943 Microsoft
$332 34,765,982 Microsoft Measurements made on past 5 days of raw data:
$16 67,808,039 Ford
$16 64,882,295 Ford Company Mean Price σ Price Sum Volume
$16 96,094,256 Ford Google $2,877 $77 8,128,196
$17 215,237,578 Ford Raw Data ==>> Amazon $3,382 $41 19,821,332
$17 100,560,723 Ford Measurements for the 5 elements Microsoft $319 $9 159,314,433
$202 3,294,689 Caterpillar Ford $16 $1 544,582,891
$200 2,933,573 Caterpillar Caterpillar $201 $3 18,389,846
$196 3,406,655 Caterpillar

Section01-ESA.xlsx - Data & Table Page 1 of 5

 Categorical data = data grouped by a category Quantitative data = numeric data
Nominal: Ordinal: Interval: Ratio:
Category Category Rank Rank
No Rank Rank Know distance between each rank Know distance between each rank
Calcs: Don't know distance Zero is either: Scales where zero = nothing exists
Count between each rank Not in the scale (like IQ scores) Ratio Okay
% of total based on the counts Calcs: Just point on the scale (like Celsius) Calcs:
Count Calcs: Count
% of total based on the counts Count % of total based on the counts
Average (Mean) OK for number category % of total based on the counts Averages
Averages Differences
Differences Ratio OK
Ratio Not Okay
Rank In Class Rating of Teacher Rating of Teacher Temperature SAT Score (400 to Score On Final Temperature
Student Phone Eye Color (1st to 4th) (Bad, Good, Great) (1,2,3) Fahrenheit 1200) (0 - 100) Kelvin Money In Bank
Chantel iPhone Brown 1st Good 2 -10 1170 81 249.8 $12,000.00
Michael Samsung Brown 4th Great 3 0 590 90 255.4 $6,000.00
Mo iPhone Brown 3th Bad 1 20 1180 45 266.5 $4,369.00
Sioux iPhone Hazel 2nd Good 2 10 1099 84 260.9 $3,131.00

Section01-ESA.xlsx - Data Terms Page 2 of 5

Types of numbers:
Discrete numbers are counting numbers & have gaps between each successive number,
like 1, 2, 3…, or 1.2, 1.3, 1.4…. Answers the question How Many?
Continuous numbers can occupy any value over a continuous range and depend on the measuring instrument.
Like: time, weight, temperature, money ($ don't seem continuous, but many statisticians treat it as such).
Answers the question How Much?

Cross-sectional data = Data collected at the same or nearly the same point in time
This cross-sectional data was collected on 11/5/2021:
Useful to compare elements such as company.

Company Ticker Year Incorporated Industry No. Employees P/E Ratio ($)
Google GOOG 2015 Software 150,028 28.12
Amazon AMZN 1996 Retail 1,335,000 58.72
Microsoft MSFT 1993 Software 181,000 36.84
Ford F 1919 Auto 186,000 20.09
Caterpillar CAT 1986 Machinery 97,300 21.82

Time-series data = Data collected over time

This data was collected for Amazon over a 5-day period
Useful for seeing trends in the past that may help us to estimate what trends there may be in the near future

Amazon Close Price,

Amazon10/25/21
ClosetoPrice,
10/29/21
10/25/21 to 10/29/21
$3,500
Date Close Price $3,447
$3,450
10/25/2021 $3,320
$3,392
$3,400 $3,376 $3,372
10/26/2021 $3,376
$3,350 $3,320 10/27/2021 $3,392
$3,300 10/28/2021 $3,447
10/25/2021 10/26/2021
10/29/2021 10/27/2021
$3,372 10/28/2021 10/29/2021

Section01-ESA.xlsx - Data Terms Page 3 of 5

Different classifications of data in statistics:

Categorical data = data grouped by a category (also known as Qualitative data)

Nominal = data with no rank or order. Like eye color: Brown, Hazel, Brown; or Phone name: iPhone, Samsung, iPhone. With this data you can
count and then calculate the % of total based on the counts. If you use a number as a category label, you cannot do arithmetic (+, -, *. /, ^) on it
- you can only count.
Ordinal = data with rank but you don't know distance between each rank, Like: Bad, Good, Great; or *, **, ***, ****; or 1st, 2nd, 3rd. You
usually count how many are in each category and then calculate the % of total. In some cases when a number is used to represent the
category, you can calculate an average (mean).
Quantitative data = numeric data

Interval = data with rank and a fixed distance between each rank, but where zero is either not in the scale (like IQ or SAT scores) or zero is just
a point on the scale (Like Fahrenheit temperature or Celsius temperature). Calculations like counting, % of count total, differences, averages
and differences are OK, but ratios between two numbers is not OK.

Ratio = data with rank and a fixed distance between each rank and zero means nothing exists. Like money, weight, height, time, Kelvin
temperature. All are scales where zero means nothing exists. Calculations like counting, % of count total, differences and ratios are OK.
Types of numbers:
Discrete numbers are counting numbers & have gaps between each successive number, like 1, 2, 3…, or 1.2, 1.3, 1.4…. Answers the question
How Many?
Continuous numbers can occupy any value over a continuous range and depend on the measuring instrument. Like: time, weight, temperature,
money ($ don't seem continuous, but many statisticians treat it as such).
Answers the question How Much?

Section01-ESA.xlsx - Data Terms Page 4 of 5

 Statistics
Numerical facts like:
USA unemployment rate reported Jan. 2021 was 6.3%
Sioux Radcoolinator student ranked at the 90th percentile for the test
3rd quarter YouTube advertising revenue was $7.20 billion vs. $7.4 billion expected.
Subject of Statistics defined:
Statistics is the art and science of collecting, analyzing, presenting and interpreting data.
Descriptive Statistics:
Data that is summarized and presented
Tabular: table of information
Graphical: charts, graphs, visualizations
Numerical: like an average (mean), median, mode
Inferential Statistics:
The process of using data obtained from a sample to make estimates and test hypotheses about the characteristics/attribute of a population
Take a sample from the population and draw reasonable conclusions that can help to estimate the unknown future.
Define Terms:
Population: The set of all elements of interest in a particular study
(In many situations it is too costly to get data from all the elements in the population)
Example ==>> Census: Collecting data for a population
Sample: A subset of the population
Example ==>> Sample survey: Collecting data for a sample

Section01-ESA.xlsx - Statistics Page 5 of 5