Business Analytics Lec. 1, 2, 3, 4, 5, 6

Business Analytics

A process that examines data using a variety of statistical tools and involves the discovery, interpretation, communication, and application of meaningful patterns in data toward effective decision making in the business workplace 

Descriptive Analytics

past data used to make a prediction

Predictive Analytics

what will happen

Prescriptive Analytics

what should we do

Step 1

Must have something to investigate/ what is the problem

Step 2

data management

Step 3

Descriptive analytics / what has happened

Step 4

Predictive analytics / what will happen

Step 5

Prescriptive Analytics / how can I make it happen

Step 6

Actionable business decisions

Step 7

Ask the next question

What are the 2 data types?

quantitative and qualitative

Is making money the business problem

no

Critical Thinking

examine, interpret, and evaluate data to make connections

Data Literacy

understand data types and data collection methods; be able to accurately communicate about data in context

Quantitative Variables

numerical data that can be measured; mathematical operations (add, average)

Discrete Variable

whole numbers

Continuous Variable

decimal, fractions

Qualitative Variables

categorical; non-numerical, can’t do calculations

Numbers that are qualitative

tag numbers, SSN, phone numbers

Ordinal

can be put into order, order matters

Nominal

no order

As reported

the context of how we use data or information

Point of sales system

accept payments from customers; keep track of sales and inventory for online and/or physical stores

Click stream data

the pathway a user takes through their online journey; how the user navigates online from search to purchase

Social Media data

computer-based sharing of ideas, thoughts, and information through virtual networks

Sensor Data

a device that detects the change in the environment and converts change into a measurable digital signal for reading or further processing

Confidential Data

identification is coded but known by someone in charge

Anonymous Data

No identification known by anyone

Population

the entire group of individuals from which we want information

Census

have data for all of the population

Sample

a part of the population we actually collect as the dataset to analyze

Elements

these are the people or things being sampled

Random Sampling

choosing a sample of data from the population such that everyone in the population has an equal chance of being chosen as part of the sample

Biased

if subjects are chosen to favor certain outcomes

Conveniant Sampling method

select a sample of people because it is convenient or easy to get the data, readily available data

Voluntary Sampling methods

self-selected sample, those who want to participate, do so. Usually participate because they have a strong opinion on the issue

Undercoverage

sampling that is incorrect because not all the population was ever on the “list of consideration” to be selected

Stratified Random Sample

divide population into non-overlapping groups then select a random sample from each strata

Systematic Sampling

list population, select random starting point, sample each nth element

Advantages of stratified sampling

improved over simple random sampling, use with cluster

Disadvantages of Stratified Sample

don’t know that strata exist in the population; therefore, requires a trial to find the strata

Advantages of Systematic Sampling

used when population is not known, starting point is moving, multiple samples from one place to

Disadvantages of Stratified Sampling

Pattern is not known to exist, random starting number is always 1 and skip is arbitrarily selected

Disadvantages of Multi-stage Cluster Sampling

Requires at least 2 steps, loses precision

Multi-stage cluster Sampling

method used to draw a sample from a population by progressively breaking down the population into smaller and smaller groups at each stage

Advantages of multi-stage cluster sampling

large geographical areas, need a large sample size

Sampling Error

wrong sampling method is used

Coverage Bias

sample does not include representation from the entire population; common when sampling nation or worldwide; using biased sampling methods

Measurement Error

sampling procedures result in collecting data that does not answer the business problem

Errors of Observation

data entered incorrectly

Undercoverage

sampling from a list that is incomplete because not all the population is on the list

No response

those who don’t respond may differ from those who do respond

Respone Bias

using leading questions or poor wording of survey; questions that misguide responses

Content Analysis

analyzing qualitative data

What analysis can lead to errors of observation?

Content analysis

Can an email be anonymous?

no

What must participant consent have?

purpose, what you’re expected to do, if the data is confidential or anonymous,

Disadvantages of using surveys

low or no responses, cost & time consuming, leads to measurement error, must get participant consent to use data