Significance of Chronbach Alpha

To fulfill my objective of having an MBA degree, I had to do a lot of research. However impractical it may seem but my college demands 4-5 research projects to be done either as assignments or for the fulfillment of course in duration of 2 years. The quality of my research might not have been at fullest level of professionalism, partly because of research bias in data collection. But still, I have learned the most adored skill of data analysis.

I enjoy analyzing data and make meaningful conclusion out of it, but for some people it might be a daunting task to do. So, I would like to put some things in most simplistic matter to help you learn more about it. Here I am trying to explain a reliability measure, Chronbach Alpha in simple terms.

What is Chronbach’s Alpha?

Chronbach’s Alpha is one of the most widely used test to check the internal consistency of any data collected for marketing research, this is an extension to “KR-20” test which is used over dichotomous items for testing internal consistency or reliability of data collected.

What does it measure?
Chronbach’s Alpha measures the similarity of responses to individual items measuring same construct in a scale for a given case. This might be a difficult sentence to digest, so I would like to put up an example in the end to show, What it measures and how to draw conclusions from its value.

What is the significance of value of Chronbach’s Alpha?
Value of Chronbach’s Alpha is generally between 0 to 1, it might be negative sometimes. But it happens generally when the data is not re-coded properly. Thus, it is absolutely important to re-code the data measuring a particular construct and then apply the reliability test (Cronbach Alpha). Further, it should be only applied to variables measuring a particular construct. There is no use of applying it to demographics data of sample.
Values of Alpha close to 1 means there is more similar responses to various items in questionnaire. However, values close to 0 means there is very less correlation among responses of individual questions. This can be better explained through an example as shown below-

Example-
In the end, I would like to share a simple example using SPSS for data analysis, to show how Chronbach’s Alpha is calculated and interpreted. Here I first create a sample file in excel to be used for Reliability analysis in SPSS.

Let me know if this explains everything through you comments, also post some suggestions/corrections (if required).

Tricks for fast Addition of numbers

Here is another post on basic quantitative aptitude techniques, particularly concerning shortcut techniques in integer  addition Some of these are very easy to follow, still if you are not able to get something, please use the comments box to start the discussion on that particular topic.

Trick 1 : Adding Consecutive Numbers

Any set of consecutive numbers can be added by this simple relation-
Sum = (Last Number – First Number + 1) * (First Number + Last Number)/2

For example, you need to add numbers from 30 to 40, then…
Using the above relation, we get, Sum = (40-30+1) * (40+30)/2 = 11*70/2 = 385

Trick 2 : Sum of Odd numbers from 1 to n

This one is very easy. Sum of odd numbers from 1 to n is ((n+1)/2)2.
For example, 1+3+5+7+9+11+13+15+17+19+21 = ((21+1)/2)2 = 112

Trick 3 : Sum of Even numbers from 2 to n

To find this, use this relation-
Sum = (n/2)*((n/2)+1)
For example, to find sum up to 100, use this relation-
Sum = (100/2) * ((100/2)+1) = 50 * 51 = 2550

However these tricks are pretty obvious and very small, but when used in competitive exams like CAT and UPSC, these can be valuable resource in reducing the time required per question.

Check out a few more related posts here-

  1. Quantitative Aptitude Techniques
  2. Syllogism Cheat Sheet

Getting Started with Oracle Certified Java Associate – OCA

So, as I was preparing for Java for a while, I thought it would be better to get certified by the developers of Java. So, I’ve decided to get certified with Oracle’s latest Java Certification Exam, starting with Oracle Certified Java Associate level exam. The information that I got to know about this exam so far is that this exam is conducted at Pearson Testing Center and costs about Rs 9000 including taxes. Anyone, who is serious about Java should take this exam, as it would give them a confidence over their programming skill.

If you are interested in giving this exam, you can see some sample questions and the syllabus of this exam at this link-

Download

I would also recommend using Enthuware Test series, once you feel you are fully prepared for exam.

Syllogism Cheat Sheet

Dealing with syllogism had always been a big problem to me. Even if I do learn it how to deal with it, I often forget how to use them when required, I guess it needs constant revision and great logical and deductive reasoning capabilities to deal with.

Logical deductions

To make things easier, I keep this small cheat sheet which contain basic formulas to deal with syllogism in verbal reasoning. I will try to explain here, how to use this cheat sheet to solve the problems which often appears in competitive exams like CAT, XAT, IIFT, and nowadays even CSAT and other job oriented exams are also looking for reasoning capabilities in candidate.


First Premise Second Premise Conclusion
All All All
All No No
All Some No Conclusion
Some All Some
Some No Some Not
Some Some No Conclusion
No All “Some Not” – Reversed
No Some “Some Not” – Reversed
No No No Conclusion
“Some Not” or “Some Not” – Reversed Anything No Conclusion


You can use above table for your reference while solving problem on syllogism. Here is an example for same-

[Q] Premise Statements-
[a]. All cities are town.
[b]. Some cities are villages.
Conclusion Options-
[i]. All villages are town.
[ii]. No village is a town.
[iii]. Some villages are town.

Answer Options –

  1. Only conclusion [i] follows.
  2. Only conclusion [ii] follows.
  3. Only conclusion [iii] follows.
  4. None of these.

This is clearly a combination of All + Some in premise statements. Hence no conclusion can be drawn. So, option 4 is correct. None of these.

I had took this question from Combined Graduate Level Exam conducted by SSC in 2013, questions like this could be a matter of 2 or 3 seconds if you can remember this table. Here is another quick example of how to use this cheat sheet.

[Q] Premise Statements
(a). All papers are books.
(b). All books are pages
(c). All pages are material.
Conclusions Options:
I. Some material are pages.
II. All books are material.
III. All papers are pages.
IV. Some books are papers.

Answer Options –

  1. All the four follow.
  2. Only II, III follow.
  3. Only I, III and IV follow.
  4. Either I or III and II follow.
  5. None follows

This one is a bit tricky question. While it seems All + All = All, should give us II & III as correct statements, but we should also look what other statements are telling us. Conclusion statement I & IV are mere implications of premise statement (a) and (c). Hence the correct answer option is [1.] All the four follows.

And last but not the least, a very simple problem, which would generally come in medium difficulty level exam-

[Q]. Premise Statements-
Statement 1. Some dogs are bats.
Statement 2. All bats are cats.

You probably don’t need options for problem like this one, as you can refer the table above, which says Some + All = Some. So, cancelling out bats, we get Some dogs are cats as our answer statement…

There are plenty of various types of questions which one can practice based on Syllogism and deductive reasoning these generally include conditional statements, possibility cases, premise – conclusion etc. Sometimes conventional way of using Venn Diagram might help in solving Syllogism but not always a good practice in competitive exams. It would always be better if you could remember the various possibility cases, and arrive to conclusion in seconds instead of working out a problem in minutes. An advanced level cheat sheet for syllogism is available here.