Science/NSF report

[Loet Leydesdorff]

Dear colleagues, 
 
In summary, the ANOVA results of a few days ago compared the core-SCI+SSCI
set (5000+) with the journals added for the SCI-E (2000+). These two groups
were significantly different on all citation indicators. The Indicators are
based on the CD-Rom versions and this makes perfect sense because it allows
the analysts to run jobs in batch mode.
 
One issue remains and that is the special status of the 400+ journals
indicated as "Professional Fields" and not included in the analysis for the
Indicators Reports. These include journals from law, business and management
studies, education, LIS (library studies), etc. The group is also
significantly different on all citation indicators from the other group of
journals (but the significances may be a result of the large numbers). 
 

 
Descriptives

 


 

 

N

Mean

Std. Deviation

Std. Error

95% Confidence Interval for Mean

Minimum

Maximum


 

 

Lower Bound

Upper Bound

Lower Bound

Upper Bound

Lower Bound

Upper Bound

Lower Bound

Upper Bound


IMPACT

.00

439

.88503

.813250

.038814

.80874

.96131

.000

6.035


 

1.00

5026

1.92364

2.837329

.040022

1.84518

2.00210

.000

49.794


 

Total

5465

1.84021

2.745242

.037135

1.76741

1.91301

.000

49.794


IMMEDIACY

.00

422

.21276

.326727

.015905

.18150

.24403

.000

2.267


 

1.00

4908

.36026

.695697

.009930

.34079

.37972

.000

21.300


 

Total

5330

.34858

.675045

.009246

.33045

.36671

.000

21.300


CITEDHALFLIFE

.00

373

27.98150

38.853235

2.011744

24.02568

31.93732

2.700

99.900


 

1.00

4800

26.70940

38.530577

.556141

25.61910

27.79969

.500

99.900


 

Total

5173

26.80112

38.551556

.536007

25.75032

27.85192

.500

99.900


Total Cites

.00

439

674.48

1173.454

56.006

564.40

784.55

3

8685


 

1.00

5049

4410.54

15377.956

216.419

3986.26

4834.81

2

404397


 

Total

5488

4111.68

14788.462

199.625

3720.34

4503.03

2

404397


Self-Citation

.00

439

80.41

125.267

5.979

68.66

92.16

0

1120


 

1.00

5049

409.50

1648.490

23.200

364.02

454.98

0

48558


 

Total

5488

383.18

1584.084

21.383

341.26

425.10

0

48558

 

Among these journals are surprises like Communication Research which is a
leading journal in the field of communication studies. In terms of
parametric statistics, it would probably not make a difference. However,
because all underlying distributions are the subject of study and highly
skewed, small differences in choices may have large impacts. 

 

The different sets are not the same, and rankings may therefore easily
differ among them. This is an empirical question.

 

With best wishes, 

 

 

Loet

  _____  

Loet Leydesdorff 
Amsterdam School of Communications Research (ASCoR), 
Kloveniersburgwal 48, 1012 CX Amsterdam. 
Tel.: +31-20- 525 6598; fax: +31-20- 525 3681 
 <mailto:loet at leydesdorff.net> loet at leydesdorff.net ;
<http://www.leydesdorff.net/> http://www.leydesdorff.net/ 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.asis.org/pipermail/sigmetrics/attachments/20080220/2fb79ede/attachment.html>