Data Business Decision Lesson 01 Role of Data

Data & Business Decision (Lesson - 01) Role of Data in modern Organization Dr. C. Ertuna 1

Management by Fact • As a manager you observed a decline in Stockholder satisfaction. (the Business Situation) • You believe that the Stockholder satisfaction is a function of Revenues, Profits Before Taxes, Return on Assets and Earnings per Share. (the Model, and the variables) • You analyze the data on these variables and decide on increasing Earnings per Share. (the Decision) • You are managing by “Fact” Dr. C. Ertuna 2

Management by Fact Managing based on models that explain the relationship between – the Business Situation and – Decision is called “Management by Fact. ” Dr. C. Ertuna 3

Management by Fact Tools • Data Analysis* • Decision Modeling * One of the most important tool for data analysis is Statistics. Dr. C. Ertuna 4

Role of Data Analysis in Business is to extract larger meaning from data in order to support – Evaluation and – Decision-making. Dr. C. Ertuna 5

Types of Business Data • • • Customer Satisfaction Data. End-User Satisfaction, etc. Financial Data EPS, ROI, etc. Market Performance Data Market share, Sales, etc Human Resources Data Employee satisfaction etc Supplier and Partner Performance Data Defects, etc. Organizational Effectiveness Data. Response Time, etc. Dr. C. Ertuna 6

Statistics - Definition - Role - Types is science of – Collecting, – Organizing, – Interpreting, and – Presenting data. Dr. C. Ertuna 7

Statistics - Definition - Role - Types • to Draw Inferences (Increases in Sales are not stable) • to Monitor the Effectiveness of business processes. (Customer Satisfaction: Customer Satisfaction Index) Dr. C. Ertuna 8

Statistics - Definition - Role - Types • Descriptive Statistics (Gender distribution of customers) • Statistical Inference (Impact of age on color preference) • Predictive Statistics. (Number of expected visitors next year) Dr. C. Ertuna 9

Decision Model - Definition - Role - Types is – Logical or – Mathematical representation of a business situation (or problem) Room_Sales = f(Room_Price, Advertisement, Service_Quality, etc. ). Dr. C. Ertuna 10

Decision Model - Definition - Role - Types is to establish a relationship between – Actions and – Results. ( 5% ↓ Room_Rates → 3% ↑ Room_Sales ) Dr. C. Ertuna 11

Decision Model - Definition - Role - Types Regression Models Forecasting Models Selection Models Simulation Models Optimization Models. Dr. C. Ertuna 12

Data Scales Data scales can be defined two ways: a) In terms of quality of data scale, such as Ratio, Interval, Ordinal, Nominal b) In terms of time of data scale, such as time series , cross-sectional Dr. C. Ertuna 13

Scales of Measurement • Ratio ( natural zero, fixed unite of measure, Sales) • Interval (no natural zero, fixed unite of measure, C) • Ordinal (no fixed unite of measurement, ranking, Priorities) • Categorical / Nominal (no ranking, Religion) No meaningful comparison of ranges, averages, and other statistics. Dr. C. Ertuna 14

Scales of Measurement • Ratio • Interval • Ordinal – natural zero, constant scale, ranked – e. g. , number of rooms sales, length of stay, age of guests, – no natural zero but constant scale, ranked, – differences make sense, but ratios do not (e. g. , 30°-20°=20°-10°, but 20°/10° is not twice as hot! – e. g. , temperature (C, F), longitudes, dates – – No constant scale, ranked but differences between values are not important e. g. , hotel ratings Surway data although ordinal could be treated like interval e. g. , degree of guest satisfaction; Likert scales, rank on a scale of 1. . 7 – – classification data, e. g. m/f, no ranking, e. g. it makes no sense to state that M > F arbitrary labels, e. g. , m/f, 0/1, etc also called count data since only numerical aspect is the number of observations • Categorical / Nominal Dr. C. Ertuna 15

Time Scale of Data • Time Series – Different dates different observation. Room sales of one hotel over many months • Ctross Sectional – On one date there are many observation Room sales of many hotels for a particular month. Dr. C. Ertuna 16

Time Series Data means that each observation is link to one unique time period different from the rest. • One date - one observation • The sequence of the data cannot be changed. Dr. C. Ertuna 17

Cross-sectional Data Cross Sectional Data means that there is one time period at which many observations are recorder. • One date - many observations • The sequence of the data can be changed. Dr. C. Ertuna 18

Data Scales in SPSS • Scale Data (both Ratio or Interval data are grouped under this name) • • Ordinal Data Categorical Data • Numeric Data (Scale & Ordinal data) • String Data (Categorical data) Dr. C. Ertuna 19

Example: Types of Data Scales • Trees are surveyed and their species and height are measured. The levels of measurement for the two variables (species and height) are; a Nominal (Categorical) and Interval b Ratio and Interval c Categorical (Nominal) and Ratio d Ordinal and Full Dr. C. Ertuna 20

Example: Types of Data Scales (cont. ) • Which of the following variables is measured at the Ordinal level? a Mean annual rainfall (e. g. millimeters) b Town Post-codes (e. g. 2840) c Accommodation ratings (e. g. 5 star, 4 star) d Temperature (e. g. degrees Celsius) Dr. C. Ertuna 21

Example: Types of Data Scales (cont. ) • Which of the following is NOT measured at the Nominal level of measurement? a Eye Color b Income Tax Bracket c Country of birth d Post-code Dr. C. Ertuna 22

How Data are used in Evaluating and Solving Business Problems Role of Data Analysis in Business Is to extract larger meaning from data to support evaluation and decision making. One of the most important tool for data analysis is statistics. Role of Statistics in Business Is to draw inferences and to monitor the effectiveness of business processes. Statistics is science of Collecting, Organizing, Interpreting, and Presenting data. Role of Decision Modeling in Business Is to establish a relationship between actions and results. Decision Model is logical or mathematical relation of a business situation (or problem). Regression Models Customer Satisfaction Data Market Performance Data Descriptive Statistics Human Resources Data Predictive Statistics Statistical Inference Supplier Performance Data Forecasting Models Selection Models Simulation Models Optimization Models Effectiveness Data Dr. C. Ertuna 23

Next Lesson (Lesson - 02) Displaying & Summarizing Data Dr. C. Ertuna 24