Package 'siie'

Title: Superior Identification Index and Its Extensions
Description: Calculate superior identification index and its extensions. Measure the performance of journals based on how well they could identify the top papers by any index (e.g. citation indices) according to Huang & Yang. (2022) <doi:10.1007/s11192-022-04372-z>. These methods could be extended to evaluate other entities such as institutes, countries, etc.
Authors: Tian-Yuan Huang [aut, cre]
Maintainer: Tian-Yuan Huang <[email protected]>
License: MIT + file LICENSE
Version: 0.4.0
Built: 2024-11-27 06:44:17 UTC
Source: CRAN

Help Index


Get data for p-SIE curve

Description

Calculate data for p-SIE curve.

Usage

p_sie(df, group, index, p = 1:100, to_compare)

Arguments

df

A data.frame containing at least two columns (namely the group and the index of each individual).

group

The group avariable. In the context of our paper, this could be the name or ISSN of a journal.

index

The indicator of individuals. In the context of our paper, this could be citation index of papers.

p

A group of p (Cutoff of superior). Defaults to integers from 1 to 100.

to_compare

Which groups to compare with.

Value

A data.table with 3 columns, with the group, p and the according SIE.

References

Huang, TY., Yang, L. Superior identification index: Quantifying the capability of academic journals to recognize good research. Scientometrics 127, 4023–4043 (2022). https://doi.org/10.1007/s11192-022-04372-z

Examples

library(ggplot2)
library(tidyfst)

set.seed(19960822)
nr_of_rows = 1e4
data.frame(
  Id = 1:1e4,
  Journal = sample(LETTERS,nr_of_rows,replace = TRUE),
  CiteCount = sample(1:100,nr_of_rows,replace = TRUE)
) -> journal_table

p_sie(journal_table,group = "Journal",
      index = "CiteCount",to_compare = c("A","B","C")) -> p_sie_df

p_sie_df

p_sie_df %>%
  ggplot(aes(p/100,sie,color = Journal)) +
  geom_point() +
  geom_line() +
  geom_abline(slope = 1,linetype = "dashed") +
  scale_x_continuous(labels = tidyfst::percent) +
  scale_y_continuous(labels = tidyfst::percent) +
  labs(x = "p",y = "SIE") +
  theme_bw()

Calculation of Paper Rank Percentile (PRP)

Description

Paper rank percentile refers to the journals' average ranking of papers within the field. If Journal A has a PRP of 90, it means the papers in Journal A has an average ranking of 90 out of 100. This metric could be extended to measure other entities such as institutes and countries.

Usage

prp(df, group, index)

Arguments

df

A data.frame containing at least two columns (namely the group and the index of each individual).

group

The group avariable. In the context of our paper, this could be the name or ISSN of a journal.

index

The indicator of individuals. In the context of our paper, this could be citation index of papers.

Value

A data.table with 3 columns, with the group, total number in the group (total_no) and PRP.

References

Huang, TY., Yang, L. Superior identification index: Quantifying the capability of academic journals to recognize good research. Scientometrics 127, 4023–4043 (2022). https://doi.org/10.1007/s11192-022-04372-z

Examples

set.seed(19960822)
 nr_of_rows = 1e4
 data.frame(
   Id = 1:1e4,
   Journal = sample(LETTERS,nr_of_rows,replace = TRUE),
   CiteCount = sample(1:100,nr_of_rows,replace = TRUE)
 ) -> journal_table

 prp(journal_table,group = "Journal",index = "CiteCount")

Calculation of SII and SIE

Description

Calculate SII (Superior Identification Index) and SIE (Superior Identification Efficiency) for each group with individual values. In the context of the paper, we have citation counts of papers from different journals. This function could calculate SII and SIE for each journal within the field.

Usage

siie(df, group, index, p = 10)

Arguments

df

A data.frame containing at least two columns (namely the group and the index of each individual).

group

The group avariable. In the context of our paper, this could be the name or ISSN of a journal.

index

The indicator of individuals. In the context of our paper, this could be citation index of papers.

p

Cutoff of superior. Defaults to 10, meaning top 10 percent individuals are regarded as superior.

Details

In the context, SII indicates how well a journal could identify the top papers (superior research), whereas SIE quantifies the efficiency of a journal to identify the top papers.

Value

A data.table with 5 columns, with the group, superior number (superior_no), total number in the group (total_no), SII (sii) and SIE (sie).

References

Huang, TY., Yang, L. Superior identification index: Quantifying the capability of academic journals to recognize good research. Scientometrics 127, 4023–4043 (2022). https://doi.org/10.1007/s11192-022-04372-z

Examples

set.seed(19960822)
nr_of_rows = 1e4
data.frame(
  Id = 1:1e4,
  Journal = sample(LETTERS,nr_of_rows,replace = TRUE),
  CiteCount = sample(1:100,nr_of_rows,replace = TRUE)
) -> journal_table
siie(journal_table,group = "Journal",index = "CiteCount")