Skip to contents

This dataset gives information about metadata coverage per year and agreement.

Usage

cr_md

Format

An object of class tbl_df (inherits from tbl, data.frame) with 84833 rows and 9 columns.

Details

The following Crossref metadata were analysed:

cr_year

Earliest publication year (Crossref field `issued`)

issn_l

Linking ISSN

articles_total

Yearly journal output by year and country affiliation

tdm_total

The number of articles containing full text URLs in the metadata

orcid_total

The number of articles containing at least one ORCID in the metadata

funder_total

The number of articles with funding metadata

abstract_total

The number of articles with open abstracts

ref_total

The number of articles with open reference lists)

cat

Global output or from first-authors based in Germany

Examples

# OA Articles in Scientometrics from first-authors in Germany: Metadata coverage
cr_md[cr_md$issn_l == "0138-9130" & cr_md$cat == "Germany",]
#> # A tibble: 8 × 9
#>   cr_year issn_l article_total tdm_total orcid_total funder_total abstract_total
#>     <int> <chr>          <int>     <int>       <int>        <int>          <int>
#> 1    2017 0138-…             7         7           3            4              0
#> 2    2018 0138-…             5         5           4            1              0
#> 3    2019 0138-…             6         6           5            1              2
#> 4    2020 0138-…            28        28          25           11             28
#> 5    2021 0138-…            28        28          25           25             24
#> 6    2022 0138-…            23        23          22           22             22
#> 7    2023 0138-…            15        15          15           15             15
#> 8    2024 0138-…            10        10          10           10             10
#> # ℹ 2 more variables: ref_total <int>, cat <chr>