Page 147 - DLIS401_METHODOLOGY_OF_RESEARCH_AND_STATISTICAL_TECHNIQUES
P. 147
Methodology of Research and Statistical Techniques
Notes such as data warehousing, executive information systems, data visualization, application
development, etc. besides statistical analysis.
(f) We have a staff of highly experienced, dedicated SAS programmers around the world.
(g) There is a long term commitment to dedicate staff to the SAS environment. There is
probably nothing, which cannot be done with SAS but getting it done is always very
difficult. There is so much literature in it that manuals take up entire shelf! May be that’s
why there are a lot of jobs out there for SAS programmers!
(h) It is expensive (unless you get an academic discount). SAS has this trick. They sell you
base SAS, which is quite limited. Then, they sell you SAS/STAT, and other modules.
These prices add up.
(i) SAS requires a lot of support, by someone who knows computers very well and knows
SAS very well. It is a pain to learn. This is partially mitigated by the fact that SAS is very
modular. Basically, there is the data step, where SAS data sets are made, manipulated
and saved. Then there is a long list of procedures (called PROC’s), which can be learned
one at a time, for the most part.
(j) The graphics in SAS are horrible. They came out with SAS/GRAPH, which has (theoretically)
vast capability, but is almost impossible to use.
However, SAS appears to be more appropriate for an enterprise solution, where the data may
reside in many different formats and SAS is the tool used to get at the data and perform
Statistical analysis. Where we are looking only for a single user PC somewhat limited use
solution the other packages appear more appropriate.
Task Write an essay on data analysis using software packages.
10.2 SPSS
Some features of SPSS:
(a) SPSS is owned by the same company, which owns some other statistical package copyright
such as SYSTAT; that is, this experience of statistical programming makes it more advanced
and user friendly.
(b) SPSS owes a lot to the days when the manual (the plum one) was one of the best
introductory books around on statistics.
(c) Non programmers find it easier to use than some other statistical packages; more menu
driven versus programming.
(d) Slower performance.
(e) Tends to be used more in the professional field.
(f) Training available everywhere.
(g) It seems that SPSS is not the software of choice in industries that need data analyst. But
SPSS has its audience too. It depends on our area of expertise.
SPSS and SAS
SPSS is a nice program for doing social research. SAS is a wonderful tool for doing data
mining. They both have their place. I can tell you that in smaller databases, such as market
142 LOVELY PROFESSIONAL UNIVERSITY