As growth of hypotesis testing tools alot of confusion of using it for research related with undergraduate thesis writing. From simple regression,multiple regression, pls sem,cb sem,gesca,ols and lsdv. To usage of its softwa from spss,amos,warp pls,visual gesca and soon.
I need your opinion which one is fit for every tool and software that i mention above for writing business management research and undergraduate thesis.
SPSS probably has the best balance of ease-of-use with being able to conduct advanced statistics. The downside is that its quite expensive. There is an open-source version called PSPP. It's free, works very much like SPSS, but it's not as powerful. Even so, it can do a surprising amount. PSPP is perfect if you're sticking with multiple regression, chi-square, and one-way ANOVA. If, for example, you would like to conduct a 2 by 3 repeated-measure ANOVA, you'll find PSPP can't handle it. Another alternative is to use web-based stats calculators. Graphpad provides quite a number of well-designed tools that you can use without downloading any software. Graphpad also sells a nice software package, which is not as powerful as SPSS, but it makes beautiful graphs, something that might be especially important for business purposes. Best wishes with your software search Dony. ~ Kevin
http://graphpad.com/quickcalcs/
http://www.gnu.org/software/pspp/
Dony -
The first thing to note is that statistical software that provides access to hypothesis testing may often contribute to its misuse, as it is often not evident that an isolated p-value is basically useless and often misused. In 1991, the (US) National Academies Press published proceedings from a forum, "The Future of Statistical Software." I noted the problem back then:
https://books.google.com/books?id=L3PshaEOnB4C&pg=PA82&lpg=PA82&dq=future+of+statistical+software+Knaub&source=bl&ots=XVHm18kKGX&sig=O59uBEzYox_st-2NXVW1QVxTWbc&hl=en&sa=X&ved=0CCcQ6AEwAGoVChMIy7vK4fLAxwIVwjU-Ch3bJAR2#v=onepage&q=future%20of%20statistical%20software%20Knaub&f=false
This is because a p-value is driven (changed) by sample size:
https://www.researchgate.net/publication/262971440_Practical_Interpretation_of_Hypothesis_Tests_-_letter_to_the_editor_-_TAS
A power analysis (type II error probability), or other sensitivity analysis, is very important. So don't overrate the usefulness of such software as an automated means to an end. If you do not understand how something works - the foundations/theory - you are more likely to misuse it. Software packages that just report p-values would be problematic.
For many advanced uses of statistics in reports, I think SAS, which is expensive, or R, which is free, can probably be very helpful. But it is always important to understand what the software is doing, and the foundations of any methodology you employ.
There is a lot to be said for using a spreadsheet. As always, you have to be careful, watch the details, and remember that confirming/testing results is important. Spreadsheets may be cumbersome, but you might have to better understand the theory behind what you are doing, and may be less likely to misapply the statistics. The graphics that spreadsheets can supply are very useful and important. Kevin mentioned graphics, and I don't think you can underestimate their value. They are often far more useful in providing true insight into your data than many metrics/statistics. If you have a theory, and you want to see if your data support it, graphics can be very important. In particular, to investigate relationships between variables, scatterplots are great. So you do want to have software that can provide good graphics.
I am not certain that you will find one software package that is especially good for every use you desire. It appears that you have a wide range of needs, not all in my expertise. I have not used SPSS in a very long time, but for business use, a quick look at the internet indicates to me that Kevin's response has merit.
Best wishes - Jim
Article Practical Interpretation of Hypothesis Tests - letter to the...
I wrote the textbook of SAS, SPSS, Statistica and JMP.
From my experiance, the operation of JMP is the most easy and support many methods.
Price is cheap and you can use free test version .
I bundled my JMP text with JMP Ver.5 sold before 2003 that run now.
I think recent ver. is 14. We bought full options of SPSS and skip version -up twice.
I could not use it. Running cost is inportant.
- Badges
- Science topic