The post Discriminant Analysis Example – Measuring Customer Loyalty appeared first on Blue Finik.
]]>A national retail chain desires to build a discriminant function that would enable the firm to distinguish between normal customers (code = 1) and loyal customers (code = 2) based on the following variables:
A. Frequency of purchases in a year.
B. Average purchase amount in a year.
C. Number of years purchasing.
Historical data was collected on the above variables and consumers were classified in to Normal and Loyal customers based on the firms experience. A representative sample extract of the data collected is provided below for your analysis.
1. Build a discriminant model to distinguish between high risk and low risk customers.
2. Determine the classification accuracy of this discriminant model.
3. State the statistical significance of the discriminant function.
4. Which one of the three causative variables is the best discriminator for creditworthiness.
5. Identify a discriminant criterion that would enable the firm to classify future applicants in to high risk and low risk categories using the discriminant function. Justify your answers.
Import data from excel in to SPSS Software using the following steps:
Go to File > open > data > select xls
Once the data is imported, go to
Analyze>classify>discriminant
Click on define range, insert the group range, for the credit card example enter minimum as 1 and maximum as 2.
Transfer the remaining variables in the independent box.
Click on statistics: Select function coefficients as unstandardized
Click on classify: In the display box click on summary table and case wise results and continue
Click Ok
GET DATA /TYPE=XLSX
/FILE=’C:\national retail chain.xlsx’
/SHEET=name ‘Sheet1’
/CELLRANGE=full
/READNAMES=on
/ASSUMEDSTRWIDTH=32767.
DATASET NAME DataSet3 WINDOW=FRONT.
DISCRIMINANT
/GROUPS=loyalty(1 2)
/VARIABLES=freq avgpurc yrs
/ANALYSIS ALL
/PRIORS EQUAL
/STATISTICS=RAW TABLE
/PLOT=CASES
/CLASSIFY=NONMISSING POOLED.
Analysis:
Go to Canonical Discriminant function coefficients.
Discriminant function/Disc Model:
For the extended values, right click on the values and click on cell properties, format value and extend the box to see the extended values.
Discriminant Model: Z = .09242099 (Frequency) + 0.0000622477 (Average purchases) + 0.1398567825 (No of years purchasing) – 4.958
Classification accuracy = 94.4 %
As seen from the casewise statistics, we have a mismatch for case 6.
Statistical significance of discriminant function
This discriminant is dividing the model in 2 groups . Eigen value is used for 2 group analysis.
Goto eigenvalues table
The eigen value is 1.965 > 1, hence the groups are distinct. i.e sum of squares among > sum of squares within.
Wilk’s lambda
1 – Var (Among) / Var (Total)
i.e., Var (Within) / Var (Total)
lambda (A) = 0.337 < .5
Therefore the groups are statistically different as seen from the eigen value and wilks lambda.
The discriminator function is a practical model where as the best discriminator function is used to distinguish the highest discriminator.
Best Discriminator
Z = .777 (frequency) + .740 (Average purchase) + .227 (number of years purchasing)
The best discriminator is Frequency.
If all the z values for the discriminants belonging to same group are added, it gives us the values for normal and loyal customers. Therefore we get 1.322 as normal and 1.322 as loyal. For future cases
Discretion criterion:
If z > 0 then add the new client to group 2 (Loyal)
If z < 0 then add the new client to group 1 (Normal)
As seen from the discriminant scores the value of z for 1^{st} case is –1.611 and this is less than 0, hence the actual group (normal) and the predicted group are the same.
As seen for case 6 the z value is 0.120 > 0, if the z value is greater then 0, then the client will belong to the 2^{nd} group (loyal), but as seen from the actual and predicted group, it belongs to the group 1. Hence the mismatch of group.
This was another discriminant analysis example. Please let me know your thoughts in the comments section. Hope you enjoyed this discriminant analysis example.
The post Discriminant Analysis Example – Measuring Customer Loyalty appeared first on Blue Finik.
]]>The post Discriminant Analysis using SPSS and Excel with Example appeared first on Blue Finik.
]]>Discriminate analysis is very similar to the multiple regression technique. The form of the Discriminant function is:
Y = a + k1.x1 + k2.x2
Where,
Y is a dependent variable. It is a grouping variable, used for classifying into 2 or more groups.
X1 and X2 are Independent variables. These are continuous scale variables
K1 and K2 are Unstandardized Discriminant function coefficient.
Discriminant analysis helps build a discriminate model in the form of a linear equation. The coefficient of the equation can be used to calculate the Discriminant score(Y), for any new data points that we want to classify into one of the groups. A decision rule is formulated for this process, to determine the cut off score, which is usually the midpoint of the mean Discriminant scores of the groups. Based on the decision rule, we classify a new object into one group or the other. This is the predicted group membership for the new object.
For e.g.: if we have 3 variables – age , income, and no of year married for which have been identified to have impact on customer credit worthiness.
We may get Discriminant function as
Credit worthiness = 0.194 (Age) + 0.0095 (Income) + 0.163 (Yrs. married) – 9.06
Accuracy of the Discriminant function is given by the classification matrix, which tells what percentage of the existing data points are correctly classified by the Discriminant function. This percentage is somewhat analogous to the R2 in regression analysis. This step also helps us identify the misrepresented cases by the model, if any.
Statistical significance of the model is explained by the Wilk’s lambda value and the eigen value
When Eigen value > 1, groups are distinct, and hence the model has good discriminating power.
Wilk’s lambda is a measure of the extent of misfit of the Discriminant solution. Values of lambda range from 0 to 1. Values close to 0 indicate that groups are distinctly different.
Values close to 1 indicate that groups are overlapping.
When lambda < 0.5, solution is statistically significant and acceptable.
Best discriminator in order to determine which one of the independent variables is more important in discriminating between groups. We refer to the Discriminant function coefficient. To overcome the problem of different measurement units, we obtain the standardized Discriminant function coefficient.
The higher the standardized Discriminant coefficient of a variable, the higher is its discriminating power.
For e.g., from the standardized table, if we get
Y = 0.907 (Age) + 0.924 (Income) + 0.285 (Yrs. married);
Best discriminator is Income.
Discriminant criteria: How to classify a new credit card applicant as “high risk” or “low risk”
We determine the means for the group centroids.
This gives us a decision rule for classifying any new case. If the Discriminant score of an applicant falls to the right of the midpoint (i.e. greater than 0), we classify the applicant as “high risk”. Alternatively, if the Discriminant score falls to the left of the midpoint (i.e. less than 0), we classify the applicant as “low risk”.
Thus, Discriminant analysis helps us determine the credit worthiness (and hence classify them into groups) of any person whose age, income and years married are known. The Discriminant model provides an unbiased decision model, which is as good as the data on which it is based.
A credit card bank has been in the business for the last 14 years, during the last 2 years their repayment default has shot up considerably. Even though the bank charges a penalty interest on all late payments, this high default rate is putting a lot of pressure on the banks recovery mechanism and has now begun to impact its profitability in this activity. The problem appears to be the credit appraisal mechanism used by the bank to evaluate credit card applicants at the time of credit card allotment. Hence the bank desires to revamp its appraisal system using its past experience.
To determine this they have conducted a suitable research. Initially the variables that have an impact on consumers credit worthiness were identified, these variables were:
A. Consumers age.
B. Monthly household income.
C. No of years married.
Historical data was collected from the banks own record and consumers were classified in to two groups as follows:
A. High risk (code = 1)
B. Low Risk (code = 2)
This was done based on the banks experience with the customers during the last two years.
A data extract of a representative sample is provided below for your analysis.
1. Build a discriminant model to distinguish between high risk and low risk customers.
2. Determine the classification accuracy of this discriminant model.
3. State the statistical significance of the discriminant function.
4. Which one of the three causative variables is the best discriminator for credit worthiness.
5. Identify a discriminant criterion that would enable the firm to classify future applicants in to high risk and low risk categories using the discriminant function. Justify your answers.
Import data from excel in to SPSS Software using the following steps:
Go to File > open > data > select xls
Once the data is imported, go to
Analyze>classify>discriminant
Click on define range, insert the group range, for the credit card example enter minimum as 1 and maximum as 2.
Transfer the remaining variables in the independent box.
Click on statistics: Select function coefficients as unstandardized
Click on classify: In the display box click on summary table and case wise results and continue
Click Ok
DISCRIMINANT
/GROUPS=Risk(1 2)
/VARIABLES=Age income Yrs_married
/ANALYSIS ALL
/PRIORS EQUAL
/STATISTICS=RAW
/PLOT=CASES
/CLASSIFY=NONMISSING POOLED.
Analysis:
Go to Canonical Discriminant function coefficients.
Discriminant function/Disc Model:
For the extended values, right click on the values and click on cell properties, format value and extend the box to see the extended values.
Q1. Build a discriminant model to distinguish between high risk and low risk customers.
Discriminant Model: Z = 0.193678 (Age) + 0.00009573959 (Income) + 0.159566 (Years of Marriage) – 9.076
Q2. Determine the classification accuracy of this discriminant model.
Go to classification accuracy:
88.9% accuracy
We have 50% accuracy as given from the problem. This model gives us 88.9%
Also, as seen from the case wise statistics we have misclassified cases that include case no 7 and 18 which do not match.
Q3. State the statistical significance of the discriminant function.
This discriminant is dividing the model in 2 groups. Eigen value is used for 2 group analysis.
Go to Eigen values table
The Eigen value is 1.811 > 1, hence the groups are distinct. i.e. sum of squares among the groups > sum of squares within the groups.
Wilk’s lambda
1 – Var (Among)/ Var (Total)
i.e., Var (Within) / Var (Total)
Lambda (A) = 0.356 < .5
Therefore the groups are statistically different as seen from the Eigen and Wilk’s lambda.
Q4. Which one of the three causative variables is the best discriminator for credit worthiness.
The discriminator function is a practical model where as the best discriminator function is used to distinguish the highest discriminator.
Best Discriminator
Z = 0.907 (Age) + 0.924 (Income) + (0.285) (Years of Marriage)
The best discriminator is income.
5. Identify a discriminant criterion that would enable the firm to classify future applicants in to high risk and low risk categories using the discriminant function. Justify your answers.
If all the z values for the discriminants belonging to same group are added, it gives us the high and the low risk values. Therefore we get 1.269 as high risk and 1.269 as low risk. For future cases
Discretion criterion:
If z > 0 then add the new client to group 2 (low risk)
If z < 0 then add the new client to group 1 (high risk)
As seen from the discriminant scores the value of z for 1^{st} case is 2.809 and this is greater then 0, hence the actual group and the predicted group are the same.
As seen for case 7 the z value is 0.820 > 0, if the z value is greater then 0, then the client will belong to the 2^{nd} group, but as seen from the actual and predicted group, it belongs to the group 1. Hence the mismatch of group.
This is the procedure for discriminant analysis. I have another Discriminant Analysis example for you. How do you go about discriminant analysis today? Which techniques do you use for discriminant analysis? Do you have any queries regarding this technique? Please let me know your inputs in the comments section. Hope you enjoyed this article.
The post Discriminant Analysis using SPSS and Excel with Example appeared first on Blue Finik.
]]>The post Factor Analysis Example – Jet Airlines Case appeared first on Blue Finik.
]]>The domestic airlines industry has been witnessing intense competition during the last 3 years. This has led to most domestic carriers reducing their passenger fares. However in the last 9 months the price of ATF (aviation turbine fuel) has shot up considerably. This has compelled most carriers to increase their ticket prices. Indian Airlines did not increase their prices, given the fact that the average domestic flier is very price sensitive, this move by Indian Airlines was expected to increase their market share considerably. However this did not happen and jet continued to be the market leader. Hence, Indian Airlines desire to understand the major factors influencing consumer preferences for airlines. To determine this they have conducted a suitable research. Initially an exploratory research was conducted and a detailed set of variables that could influence choice of airlines were identified. These variables were then converted in to a questionnaire, an extract of which is provided below for your ready reference. Recent flyers of jet were intercepted on arrival and were asked to provide their responses to each statement in the questionnaire on a 7 point Likert scale where
1 – Strongly agree
7 – Strongly disagree
A data extract of this research is provided below for your analysis.
Q1. Identify the number of major factors influencing consumer behavior.
Q2. Determine the percentage of total variance in the data that is explained by the extracted factors cumulatively.
Q3. Identify the major constituent attributes of each factor.
Q4. Label each factor based on their dominant characteristics.
Q5. Draw perceptual maps with the factors as the axis taken 2 at a time.
Questionnaire extract:
1. They (JA) are always on time.
2. The seats are very comfortable.
3. I love their food
4. The air hostesses are beautiful.
5. My boss/friends fly on jet a
6. JA has younger aircraft.
7. They have a freq flyer program.
8. The flight timings suit my schedule
9. My mom.fly feels safe when I fly jet.
10. Flying jet complements my lifestyle and social standing in society.
1. Goto Data reduction
2. Click on factor
3. Transfer all the factors: The names are shortened and used in the analysis. Example: The variable “manvech” means man’s vehicle.
4. Click on extraction: Principal component
5. Click on rotation: Select Varimax
FACTOR
/VARIABLES ONTIME SEATCOM LOVEFOOD AIRHOBEA BOSSFRJE YOUAIRCR FREQFLPR SUITSCHE FEELSAFE COMLIFST
/MISSING LISTWISE
/ANALYSIS ONTIME SEATCOM LOVEFOOD AIRHOBEA BOSSFRJE YOUAIRCR FREQFLPR SUITSCHE FEELSAFE COMLIFST
/PRINT INITIAL EXTRACTION ROTATION
/CRITERIA MINEIGEN(1) ITERATE(25)
/EXTRACTION PC
/CRITERIA ITERATE(25)
/ROTATION VARIMAX
/METHOD=CORRELATION.
Highlighted correlation factors.
Highlighted correlation factors.
Highlighted correlation factors.
Here are the questions we had earlier:
Q1. Identify the number of major factors influencing consumer behaviour.
Q2. Determine the percentage of total variance in the data that is explained by the extracted factors cumulatively.
Q3. Identify the major constituent attributes of each factor.
Q4. Label each factor based on their dominant characteristics.
Q5. Draw perceptual maps with the factors as the axis taken 2 at a time.
1. Three factors influence the consumer behaviour:
a. F1 (On time, Love food, Younger aircraft) – Features
b. F2 (Air hostess beautiful, Frequent flyer program, Complements my lifestyle) – Flying Incentives
c. F3 (Seats comfortable, suits my schedule) – Convenience
2. Total variance explained is 80.720%
3. Refer to answer number 1
4. Features, Flying incentives and Convenience
5. Perceptual maps in excel.
Please let me know if this SPSS tutorial was helpful. I will love to hear from you regarding your suggestions or any new techniques you might be using for Factor Analysis. Please leave your comments below.
The post Factor Analysis Example – Jet Airlines Case appeared first on Blue Finik.
]]>The post Benefits of PPC appeared first on Blue Finik.
]]>Awareness is the key contributor for the success of any and every business. Business cannot strive without awareness. With PPC Management campaigns, companies can create awareness by targeting the right audience at the right time, using the right display platform. PPC advertisements have become an indispensable part for any on line marketing effort. Many companies have realized the Benefits of PPC campaigns as it has helped them increase traffic on their website hence leading to increased brand awareness and conversion.
Many factors are involved in building a brand. It takes a long time for a company to become a trusted and recognized brand. Gone are the days when traditional marketing mix like outdoor hoardings, banners, TV commercials and radio advertisements were enough for brand building. Today the world is replete with computers, laptops, tablets and smart phones which has changed the face of marketing. Organizations not having a website or any sort of online presence is sure to loose to its competition. In online marketing arena, companies can build brand by investing in both PPC campaigns and SEO. PPC combined with SEO helps organizations to have a balanced approach towards getting audience and ranking on the search engines. Advantages of Pay Per Click and SEO together are far more than any other marketing campaigns or marketing mix as these two platforms are getting enhanced every second. Benefits of PPC also includes reliability factor, Google penalizes fake or harmful websites or any other factors which can affect its customers interest thus ensuring quality search experience.
One of the benefits of PPC advertisement campaigns is that it helps companies to streamline their marketing efforts. Companies can leverage Pay Per Click platform to target the right audience. Your outreach is targeted towards prospects with buying intent hence you are assured optimum utilization of your budget. The Benefits of PPC campaigns is that it displays your ads in front of your prospects.
There is a good chance that your competitors are already deriving Benefits of PPC campaigns by getting more traffic and conversions on multiple parameters. You can design your campaign pragmatically by learning what your competitors are doing to get traffic and CTR (click through rates) which improves websites ranking and accesibility for end customers. Google want its customers to get quality searches for their needs, thus it supports healthy and transparent competition. Technically speaking, Pay Per Click platforms provide marketing team with necessary insight about your competitor’s strategy.
Keywords are the foundation of every PPC campaign hence every campaign is unique to the business and the target audience. I assume you already know that keywords are the written impression of the searchers needs (now you know this J ). Strategical Advantages of Pay Per Click campaigns are that these campaigns are highly flexible and can be customized to suit your marketing requirements.
You won’t have to cry your lungs out for this traffic because this is the traffic you will love and will want more of it. Pay Per Click campaign is all about relevant traffic, relevant traffic and more relevant traffic which you will get on your website (obviously if traffic police is right J )
Pay Per Click campaigns helps you to display your adds in front of people who are genuinely looking for your products or services. PPC campaigns have to be crafted very carefully to fit your business in order to get relevant traffic and avoid wastage of budget. Besides, you will soon realize whether PPC is a good fit for your business marketing strategy or not.
Now we are talking about the best feature of PPC campaigns,,, “accurate measurement” of your marketing effort. All Pay Per click platforms like Google, Facebook or LinkedIn provides you with plethora of reports which will give you accurate picture of every penny spent and the health of your PPC campaigns. This feature is by far one of the best Advantages of Pay Per Click campaigns over other marketing campaigns. All I can say that PPC advertisement is a great way to market your product or service as many organizations have already reaped Benefits of PPC campaigns and many are benefiting from it and all it requires is a little bit of faith and smart strategy and soon you will be witnessing good volume of organic traffic. PPC enabling platforms are continuously enhancing right from the granular level to improve online experience of both companies and customers.
The post Benefits of PPC appeared first on Blue Finik.
]]>The post Factor Analysis appeared first on Blue Finik.
]]>Factor analysis is a very useful method of reducing data complexity by reducing the number of variables being studied.
Brands/objects have large number of different attributes and factors that define them. Each attribute may or may not be correlated to each other. Also, certain attributes may exist in combinations/underlying dimensions and some may be unique. These combinations/underlying dimensions are called as Factors.
Factor Analysis helps the researcher to determine the important attributes or factors that affect the consumer buying behavior. Also, it reduces the complexity of the attributes by correlating/joining them into factors. The distinguishing attributes identified are grouped under the factors and the Perceptual Maps are drawn with the factors as the axes. The attributes are plotted on the perceptual maps. The direction of the vector line indicates the nature of association between the factor and the attribute and the length of the vector line determines the strength of association of the factor with the attributes.
In marketing, we need to know the exact reasons why a consumer buys the product. The consumers’ purchasing criteria varies in number from 2 or 3 to 15 or 20. We need to understand the underlying significant drivers of buying behavior for a particular product. Factor Analysis reduces the complexity of the features/attributes into relevant factors. Factors thus created provide insight into relevant psychographic of target customer.
The basic problem is identified and the attributes that define the problem are obtained through Questionnaire and Focus Group Discussions. The identified variables are then converted into questionnaire using questionnaire design techniques. The questionnaire was in the form of statements and the respondents are asked to rate them in a Likert scale ranging from 1 to 5 where 1Strongly Agree and 5Strongly Disagree.
The responses are loaded into SPSS and the variables that significantly affect the behavior are recorded. First, the total variable explained table contains all the factors (components) with their respective eigen values. The factors with eigen values > 1 are considered.
Also, from the Rotated Component Matrix, correlation between attributes and factors can be obtained wherein the attributes having r value (coefficient of determination) > 0.7 and <0.7 are related to the corresponding components (factors).
Once the factors are identified and the variables are grouped under the factors, the factors are given the suitable names. Then the attributes are plotted on the perceptual maps with the factors as the axes, taken 2 at a time. Thus, for every combination of 2 factors, the perceptual maps are drawn.
The length of the vector gives the strength of association and the direction gives the nature of the association.
A two wheeler producer desires to understand the major factors that influence consumer buying behavior for their products. For this purpose they have conducted a suitable research. Initially an exploratory research was conducted and a detailed set of variables that could influence consumer preferences were identified. These variables were then converted in to a questionnaire in the form of statements. Respondents were asked to provide their responses to each statement in the questionnaire on a seven point Likert scale where 1 stood for strongly agree and 7 for strongly disagree. The attributes were:
1. Affordability
2. Safety
3. Comfort
4. Economy
5. Friends’ jealous
6. 3 people on a ride
7. Man’s vehicle
8. Power
9. Ads feel good
10. Sense of freedom
An extract of the questionnaire along with a data extract is provided below for your analysis.
Q1. Identify the number of major factors influencing consumer behavior.
Q2. Determine the %ge of total variance in the data that is explained by the extracted factors cumulatively.
Q3. Identify the major constituent attributes of each factor.
Q4. Label each factor based on their dominant characteristics.
Q5. Draw perceptual maps with the factors as the axis taken 2 at a time.
Plot the relevant attributes on these maps as vectors appropriately.
Justify your answers.
1. I use a twowheeler because it is affordable.
2. It gives me a sense of freedom to own a twowheeler.
3. Low maintenance
cost. Makes it economical in long run.
4. Twowheeler is a man’s vehicle essentially.
5. I feel powerful on my twowheeler
6. Some of my friends who don’t have a 2 wheeler are jealous of me
7. I feel good when I see ads for my twowheeler
8. My twowheeler gives me a comfortable ride.
9. I think twowheelers are a safe way to travel.
10. 3 people should be legally allowed to travel on a two wheeler.
1. Goto Data reduction
2. Click on factor
3. Transfer all the factors: The names are shortened and used in the analysis. Example: The variable “manvech” means man’s vehicle.
4. Click on extraction: Principal component
5. Click on rotation: Select Varimax
FACTOR
/VARIABLES afford sof maitcost manvech powerful frdsjeo felgodad comfort safe legal
/MISSING LISTWISE
/ANALYSIS afford sof maitcost manvech powerful frdsjeo felgodad comfort safe legal
/PRINT INITIAL EXTRACTION ROTATION
/CRITERIA MINEIGEN(1) ITERATE(25)
/EXTRACTION PC
/CRITERIA ITERATE(25)
/ROTATION VARIMAX
/METHOD=CORRELATION.
Total Variance Explained  
Component 
Initial Eigenvalues 
Extraction Sums of Squared Loadings 
Rotation Sums of Squared Loadings 

Total 
% of Variance 
Cumulative % 
Total 
% of Variance 
Cumulative % 
Total 
% of Variance 
Cumulative % 

1 
3.883 
38.828 
38.828 
3.883 
38.828 
38.828 
3.841 
38.409 
38.409 
2 
2.777 
27.770 
66.598 
2.777 
27.770 
66.598 
2.429 
24.294 
62.703 
3 
1.375 
13.747 
80.346 
1.375 
13.747 
80.346 
1.764 
17.643 
80.346 
4 
.945 
9.449 
89.795 

5 
.479 
4.793 
94.588 

6 
.292 
2.923 
97.511 

7 
.117 
1.166 
98.677 

8 
.068 
.680 
99.356 

9 
.037 
.374 
99.730 

10 
.027 
.270 
100.000 

Extraction Method: Principal Component Analysis. 
Rotated Component Matrix^{a}  
Component 

1 
2 
3 

Affordable 
.126 
.313 
.780 
Sense of Freedom 
.181 
.639 
.107 
Maintainence Cost 
.116 
.604 
.594 
Man’s Vehicle 
.970 
.064 
.006 
Powerful 
.964 
.131 
.063 
Friends Jeolous 
.945 
.140 
.030 
Feel Good Ads 
.971 
.024 
.106 
Comfortable 
.262 
.848 
.101 
Safe 
.010 
.881 
.044 
3 People Legal 
.063 
.149 
.874 
Extraction Method: Principal Component Analysis. Rotation Method: Varimax with Kaiser Normalization. 

a. Rotation converged in 5 iterations. 
Rotated Component Matrix(a)  

Component  
1 
2 
3 

Affordable 
0.126 
0.313 
0.78 
Sense of Freedom 
0.181 
0.639 
0.107 
Maintainence Cost 
0.116 
0.604 
0.594 
Man’s Vehicle 
0.97 
0.064 
0.006 
Powerful 
0.964 
0.131 
0.063 
Friends Jeolous 
0.945 
0.14 
0.03 
Feel Good Ads 
0.971 
0.024 
0.106 
Comfortable 
0.262 
0.848 
0.101 
Safe 
0.01 
0.881 
0.044 
3 People Legal 
0.063 
0.149 
0.874 
Three factors are obtained in this example. Here are the factors with corresponding attributes:
F1: Economy (Affordability, 3 People Legal)
F2: Features (Comfort, Safety)
F3: Macho (Man’s vehicle, Power, Friends’ Jealous, Ads feel good)
Features Vs Economy 

Affordable 
0.313 
0.78 

Comfortable 
0.848 
0.101 

Safe 
0.881 
0.044 

3 People Legal 
0.149 
0.874 
Macho Image Vs Features 

Man’s Vehicle 
0.97 
0.064 
Powerful 
0.964 
0.131 
Friends Jeolous 
0.945 
0.14 
Feel Good Ads 
0.971 
0.024 
Comfortable 
0.262 
0.848 
Safe 
0.01 
0.881 
This is how you perform factor analysis using SPSS and Excel. Hope this SPSS factor analysis tutorial was helpful. Please add your comments below.
Image Credit: Paul Townsend
The post Factor Analysis appeared first on Blue Finik.
]]>