Tuesday, September 4, 2012

Day 1-Team G (2)


“All knowledge that the world has ever received comes from the mind; the infinite library of the universe is in our own mind”
Business leaders are yearning for deeper knowledge and insights on all aspects of their business and they know that the information they need is available within all the data flowing through the company IT systems.  The growth of social conversations and content across the many channels is providing business leaders with a never ending stream of incoming data.  The challenge is to turn all that data into insights and then develop strategies/actions based on those insights. That is where Business Analytics play a crucial role.
Learning’s from Business Analytics Lecture
Business Analytics’ AKA ‘Business Intelligence’ allows users to examine and manipulate data to drive positive business actions armed with advanced analytics insights, business users can make well-informed, fact-based decisions to support their organizations’ tactical and strategic goals.

The underlying objective: “Discover new and meaningful patterns in data”

But the amount of data is doubling every year. Information from devices, machines and social data, creates an entirely new set of challenges and reinforces the fact that data will continue to grow exponentially for the foreseeable future. There arises need of advanced tools and techniques. SPSS is one of them.

SPSS

“SPSS –A Social Science Statistical Package is a computer program used for:


                                         

SPSS is very similar to Microsoft Excel in Layout. There is a menu and a tool bar option at the top of every window and some of its functions are just like an Excel.

Getting Familiar with SPSS
Launching SPSS: Launch SPSS uses a start menu or a desktop shortcut. SPSS starts and the opening screen resembles a spread sheet. A dialogue box will open which will allow opening an existing data or creating a new data.
Variables and Cases: SPSS uses data organised in rows and columns. Rows are cases and columns are variables. Variables are categorised as:
 
        
Category Variables: These variables are limited in number with fixed number of possible values. For Example: Gender- There are only 2 fixed values i.e. Male and Female.
Continuous Variables: Continuous Variables can further be classified into :
  1. Continuous : Age (Years, Months and Days) e.g. John is 22 years 2 month 10 days old
  2. Discrete: These numbers have fixed values. E.g. Number of brothers and sisters.

The users of the SPSS have two views:
1. Data View: Contain data in form of number or texts.
2. Variable View: In this the user gets information about variables included in the data set. The columns provide information about the various characteristics of variables. There are ten columns in total. Each column has a different significance for variables. Columns include:
·       Name : Name of the variable
·     Type : There are 8 options which include Numeric, Comma, Dot, Scientific Notation, Date, Dollar, Custom Currency and String.
·      Width : Helps to decide number of characters to display
·      Decimal: Helps to control number of characters after the decimal place.
·      Labels: Helps to give complete description of a variable.
·     Values:Helps to provide a key for what the numbers of a numeric variable may represent. (E.g. 1= Male, 2= Female )
·   Missing: Helps to indicate missing values in the variable. These missing values are not used for analysis.
·       Columns: Helps to indicate total number of columns a variable value may have.
·       Align: Help to align data to right or left.
·      Measure: Helps to indicate the level of measurment of a variable. There are three forms :Nominal, Ordinal and Scale

Level of Measurement of Variable

Defining of data is of no use without performing operations and analysing data. There are three type of analysis based on type of data:
1. Uni-variate: Simplest form of analysing a single variable data. Data can be represented in form of pie chart.
2. Bi-variate: It includes analysis of two variables simultaneously through cross tabulation or pie charts.
3. Multivariate: It includes analysis of more than 2 variables through scatter diagram etc.
In more advanced analysis, data can be transformed from one form to the other through TRANSFORM tab. For e.g. Age from continuous discrete form can be transformed to groups.

Data can be analysed though ANALYSE tab. Under ANALYSE tab there is a descriptive option. Descriptive Statistics in SPSS include:
·     
      Frequencies: They are primarily used for discrete data (e.g., nominal and ordinal data).It tells the number of times the variable has occurred. E.g.  To know the shopping frequency of  a person. Data can be showed in terms of per cent and can be combined and analysed through cumulative per cent.

·   Cross Tabulation: They are primarily used to tabulate two or more data variables and to assess the relationship between them. Cross Tabulation generates bi-variant relation. If continuous variables are there, they are first re-categorized into category variables and then analysis is done.

e.g.  Age can be 19,20,21,22,23,24,25,26,27,28,29 and 30 years. The task is to find out how many people marry before the age of 25 years. Depicting this data in a cross tabulated form can be very tedious. Thus it can be re-categorised in to age groups 19-25 and after 25. 

Before Cross Tabulation, Hypothesis and Null Hypothesis need to be form.

Hypothesis: H1: Unemployment affects the children going to school
Null Hypothesis: H0: There is no relation between Unemployment and children going to school.

After developing Hypothesis and performing Cross Tabulation, to assess whether relationship is there between two variables or not, Test of Independence i.e. Chi-square test is performed which comes under testing of Hypothesis.
Now according to the Chi-Square test, if the significant value is lower than .05, there is a significant difference between ratios. So null hypothesis would be rejected.

Author
Akhil Aggarwal
Roll No: 14126 | SIBM (2011-13) Batch
(Team G)

No comments:

Post a Comment