targets.Rmd

---
title: "SCORE Targets"
output:
  html_document:
    toc: true
    toc_float: true
    toc_depth: 6
---

```{r setup, include = FALSE}
knitr::opts_chunk$set(collapse = TRUE, comment = "#>")
library(targets)
library(readr)
library(googledrive)
library(googlesheets4)
library(dplyr)
library(visNetwork)
tar_unscript()
```

# The Targets Package

Rendering this file using `rmarkdown::render("targets.Rmd")` executes a series of functions in the correct order to build target files (also known as "targets", e.g., analytic datasets, figures) from source files (e.g., raw data). It also documents the dependencies of the target file, including all files (whether source files or other targets) which are used as inputs to the commands in the functions used to generate the target file.

# Globals

We first define the global options and source functions used to generate target files.

```{targets globals, tar_globals = TRUE}
options(tidyverse.quiet = TRUE,
        scipen = 999)

tar_option_set(
  # Packages the targets need to run
  packages = c("BayesFactor",
               "BayesRep",
               "BFF", # Version 4.2.1 REQUIRED
               "dplyr",
               "effectsize",
               "ggExtra",
               "glue",
               "googlesheets4",
               "googledrive",
               "here",
               "kableExtra",
               "lubridate",
               "metaBMA",
               "metafor",
               "osfr",
               "purrr",
               "pwr",
               "readr",
               "ReplicationSuccess",
               "sjPlot",
               "stringr",
               "tibble",
               "tidyr") 
)

# All functions used to create targets are found in the folder named "pipeline"
tar_source(files = "pipeline")

```

# Targets

## Raw Data

Most raw data files will have three targets associated with them - one for their ID on OSF, one for the modification date of the file, and one that actually represents the data loaded into R. 

The file targets are the GUID for the file on [OSF](https://osf.io/dtzx4/files/osfstorage). These files can be found in the "Raw Data" folder.

The modification date targets are set to run every time `tar_make()` is called (with the parameter `cue = tar_cue("always")`) so that the file is pulled again if, and only if, it is ever updated online. 

The `get_osf_mod_date()` and `load_osf_csv()` functions can be found in `pipeline/helper_download_osf.R`. 

### Journal Metadata

``` {targets raw-journal}
list(
  # journal_metadata_raw.csv ----
  # Metadata about journals that articles were sourced from, including names,
  # ISSNs, and general publication category
  tar_target(journal_metadata_file,
             "tcb8q"),
  tar_target(journal_metadata_moddate,
             get_osf_mod_date(journal_metadata_file),
             cue = tar_cue("always")),
  tar_target(journal_metadata_raw,
             load_osf_csv(journal_metadata_file, 
                          journal_metadata_moddate)),
  
  # journal_repro-policies.csv ----
  # Journal policies about reproducibility and data/code sharing
  tar_target(journal_repro_policies_file,
             "24j8s"),
  tar_target(journal_repro_policies_moddate,
             get_osf_mod_date(journal_repro_policies_file),
             cue = tar_cue("always")),
  tar_target(journal_repro_policies,
             load_osf_csv(journal_repro_policies_file,
                          journal_repro_policies_moddate)),
  
  # journal_top-factor.csv ----
  # TOP Factor (2015) standards for journals that articles were sourced from,
  # according to the TOP factor website (https://topfactor.org/journals)
  tar_target(journal_top_factor_file,
             "2bhwe"),
  tar_target(journal_top_factor_moddate,
             get_osf_mod_date(journal_top_factor_file),
             cue = tar_cue("always")),
  tar_target(journal_top_factor,
             load_osf_csv(journal_top_factor_file, 
                          journal_top_factor_moddate))
)
```

### Paper Metadata

```{targets raw-paper}
list(
  # paper_citations.csv ----
  # Citations in APA format for papers from which claims were extracted
  tar_target(paper_citations_file,
             "wr8a3"),
  tar_target(paper_citations_moddate,
             get_osf_mod_date(paper_citations_file),
             cue = tar_cue("always")),
  tar_target(paper_citations,
             load_osf_csv(paper_citations_file, 
                          paper_citations_moddate)),
  
  # paper_metadata_covid.csv ----
  # Metadata for COVID preprints from which claims were extracted
  tar_target(paper_metadata_covid_file,
             "dc3uv"),
  tar_target(paper_metadata_covid_moddate,
             get_osf_mod_date(paper_metadata_covid_file),
             cue = tar_cue("always")),
  tar_target(paper_metadata_covid,
             load_osf_csv(paper_metadata_covid_file,
                          paper_metadata_covid_moddate)),
  
  # paper_metadata_raw.csv ----
  # Raw metadata for papers from which research claims were extracted 
  # (excluding COVID preprints)
  tar_target(paper_metadata_raw_file,
             "j4gkp"),
  tar_target(paper_metadata_raw_moddate,
             get_osf_mod_date(paper_metadata_raw_file),
             cue = tar_cue("always")),
  tar_target(paper_metadata_raw,
             load_osf_csv(paper_metadata_raw_file,
                          paper_metadata_raw_moddate))
)
```

### Original Studies

#### Original Study Claims

```{targets raw-claims}
list(
  # orig_claims_bushel-complex.csv ----
  # Indicates bushel claims where the evidence is anything other than a single
  # statistically significant effect
  tar_target(orig_claims_bushel_complex_file,
             "qjgwb"),
  tar_target(orig_claims_bushel_complex_moddate,
             get_osf_mod_date(orig_claims_bushel_complex_file),
             cue = tar_cue("always")),
  tar_target(orig_claims_bushel_complex,
             load_osf_csv(orig_claims_bushel_complex_file,
                          orig_claims_bushel_complex_moddate)),
  
  # orig_claims_bushel-non-sig.csv ----
  # Indicates which bushel claims use non-significant evidence.
  tar_target(orig_claims_bushel_sig_gsheet,
             "sgqf8"),
  tar_target(orig_claims_bushel_sig_moddate,
             get_osf_mod_date(orig_claims_bushel_sig_gsheet),
             cue = tar_cue("always")),
  tar_target(orig_claims_bushel_sig,
             load_osf_csv(orig_claims_bushel_sig_gsheet,
                          orig_claims_bushel_sig_moddate)),
  
  # orig_claims_covid_ta2-ids.csv ----
  # Claims extracted from phase 1 COVID preprints with IDs for human assessment
  # teams. Used to reconcile IDs for process reproducibility work.
  tar_target(orig_claims_covid_ta2_file,
             "q4kca"),
  tar_target(orig_claims_covid_ta2_moddate, 
             get_osf_mod_date(orig_claims_covid_ta2_file),
             cue = tar_cue("always")), 
  tar_target(orig_claims_covid_ta2,
             load_osf_csv(orig_claims_covid_ta2_file,
                          orig_claims_covid_ta2_moddate)),
  
  # orig_claims_p1-covid.csv ----
  # Claims extracted from phase 1 COVID preprints with canonical paper IDs.
  # Preprints were from PsyarXiv, SocArXiv, and medRxiv.
  tar_target(orig_claims_p1_covid_file,
             "94dfs"),
  tar_target(orig_claims_p1_covid_moddate,
             get_osf_mod_date(orig_claims_p1_covid_file),
             cue = tar_cue("always")),
  tar_target(orig_claims_p1_covid,
             load_osf_csv(orig_claims_p1_covid_file,
                          orig_claims_p1_covid_moddate)),
  
  # orig_claims_p1.csv ----
  # Claims extracted from articles during phase 1 (February 2019 to December
  # 2020) with canonical paper IDs
  tar_target(orig_claims_p1_file,
             "na35f"),
  tar_target(orig_claims_p1_moddate,
             get_osf_mod_date(orig_claims_p1_file),
             cue = tar_cue("always")),
  tar_target(orig_claims_p1,
             load_osf_csv(orig_claims_p1_file,
                          orig_claims_p1_moddate)),
  
  # orig_claims_p2-bushel.csv ----
  # Bushel claims from phase 2 with canonical paper IDs. 
  # Rather than focusing on an individual claim in a paper, bushel claim 
  # traces may branch at any point of a paper up until the 
  # test description (claim 4)
  tar_target(orig_claims_p2_bushel_file,
             "9m57d"),
  tar_target(orig_claims_p2_bushel_moddate,
             get_osf_mod_date(orig_claims_p2_bushel_file),
             cue = tar_cue("always")),
  tar_target(orig_claims_p2_bushel,
             load_osf_csv(orig_claims_p2_bushel_file,
                          orig_claims_p2_bushel_moddate)),
  
  # orig_claims_p2-single.csv ----
  # Single-trace claims extracted from articles during phase 2 (January 2021 to
  # May 2022) with canonical paper IDs
  tar_target(orig_claims_p2_single_file,
             "nakxy"),
  tar_target(orig_claims_p2_single_moddate,
             get_osf_mod_date(orig_claims_p2_single_file),
             cue = tar_cue("always")),
  tar_target(orig_claims_p2_single,
             load_osf_csv(orig_claims_p2_single_file,
                          orig_claims_p2_single_moddate))
)
```

#### Original Study Variables

```{targets raw-orig}
list(
  # orig_effect-size.csv ----
  # Calculated effect sizes with upper and lower bounds for original studies
  tar_target(orig_effect_size_file,
             "4s3de"),
  tar_target(orig_effect_size_moddate,
             get_osf_mod_date(orig_effect_size_file),
             cue = tar_cue("always")), 
  tar_target(orig_effect_size,
             load_osf_csv(orig_effect_size_file,
                          orig_effect_size_moddate)),
  
  # orig_n_ser.csv ----
  # Set of sample sizes calculated via the SER method for power calculations
  tar_target(orig_n_ser_file,
             "nhw8q"),
  tar_target(orig_n_ser_moddate,
             get_osf_mod_date(orig_n_ser_file),
             cue = tar_cue("always")),
  tar_target(orig_n_ser,
             load_osf_csv(orig_n_ser_file,
                          orig_n_ser_moddate)),
  
  # orig_n_traditional.csv ----
  # Set of sample sizes calculated via traditional methods for 
  # power calculations
  tar_target(orig_n_traditional_file,
             "pbftz"),
  tar_target(orig_n_traditional_moddate,
             get_osf_mod_date(orig_n_traditional_file),
             cue = tar_cue("always")), 
  tar_target(orig_n_traditional,
             load_osf_csv(orig_n_traditional_file,
                          orig_n_traditional_moddate)),
  
  # orig_stats_effective.csv ----
  # Adjusted sample sizes and degrees of freedom for the purpose of 
  # effect size estimation.
  tar_target(orig_stats_effect_file,
             "k5chq"),
  tar_target(orig_stats_effect_moddate,
             get_osf_mod_date(orig_stats_effect_file),
             cue = tar_cue("always")), 
  tar_target(orig_stats_effect,
             load_osf_csv(orig_stats_effect_file,
                          orig_stats_effect_moddate)),
  
  # orig_stats_extended.csv ----
  # Additional statistics for original studies calculated by collaborators at
  # Tilburg University.
  tar_target(orig_stats_extended_file,
             "9fbm8"),
  tar_target(orig_stats_extended_moddate, 
             get_osf_mod_date(orig_stats_extended_file),
             cue = tar_cue("always")), 
  tar_target(orig_stats_extended,
             # Delivered data has everything encased in quotes
             # These are removed before merging into analytic datasets
             load_osf_csv(orig_stats_extended_file,
                          orig_stats_extended_moddate)),
  
  # orig_stats_p1.csv ----
  # Statistics from original articles that underwent replication/reproduction 
  # in phase 1. Each record is a paper that documents the single claim and
  # inferential test selected for that paper. In Phase 1, the (unique)
  # inferential test from the (unique) focal claim is coded and stored across
  # several different workflows, then aggregated into a ground truth version
  # post QA and merging. Original variables may have been coded in either/both
  # of main or repro type pipelines, so there are sometimes multiple rows 
  # per paper.
  tar_target(orig_stats_p1_file,
             "fa6ub"),
  tar_target(orig_stats_p1_moddate,
             get_osf_mod_date(orig_stats_p1_file),
             cue = tar_cue("always")),
  tar_target(orig_stats_p1,
             load_osf_csv(orig_stats_p1_file,
                          orig_stats_p1_moddate)),
  
  # orig_stats_p2_additional.csv ----
  # Statistics from original articles that underwent replications/reproductions
  # in phase 2 that were entered manually later in the process
  tar_target(orig_stats_p2_additional_file,
             "y84gf"),
  tar_target(orig_stats_p2_additional_moddate, 
             get_osf_mod_date(orig_stats_p2_additional_file),
             cue = tar_cue("always")), 
  tar_target(
    orig_stats_p2_additional,
    load_osf_csv(
      orig_stats_p2_additional_file,
      orig_stats_p2_additional_moddate,
      col_types = list(
        # All entries "F", which should load as char, not logical
        original_statistic_type_reported = col_character(),
        original_statistic_analysis_type_statsteam = col_character()
      )
    )
  ),
  
  # orig_stats_p2_form-outputs.csv ----
  # Statistics from original articles that underwent replications/reproductions
  # in phase 2 that were entered via google form.
  tar_target(orig_stats_p2_form_file,
             "6qg4a"),
  tar_target(orig_stats_p2_form_moddate,
             get_osf_mod_date(orig_stats_p2_form_file),
             cue = tar_cue("always")), 
  tar_target(orig_stats_p2_form,
             load_osf_csv(orig_stats_p2_form_file,
                          orig_stats_p2_form_moddate)),
  
  # orig_stats_p2_manual-entry.csv ----
  # Statistics from original articles that underwent replications in phase 2
  # that were entered manually rather than through the other workflows.
  tar_target(orig_stats_p2_manual_file,
             "dusqm"),
  tar_target(orig_stats_p2_manual_moddate,
             get_osf_mod_date(orig_stats_p2_manual_file),
             cue = tar_cue("always")),
  tar_target(
    orig_stats_p2_manual,
    load_osf_csv(
      orig_stats_p2_manual_file,
      orig_stats_p2_manual_moddate,
      col_types = list(
        # Empty columns that should load as char, not logical
        original_statistic_type_reported = col_character(),
        original_statistic_type_reference = col_character()
      )
    )
  ),
  
  # orig_stats_p2_power-outputs.csv ----
  # Statistics from original articles that underwent replications in phase 2
  # that were gathered from power analysis outputs
  tar_target(orig_stats_p2_pwr_file,
             "6cweg"),
  tar_target(orig_stats_p2_pwr_moddate,
             get_osf_mod_date(orig_stats_p2_pwr_file),
             cue = tar_cue("always")),
  tar_target(orig_stats_p2_pwr,
             load_osf_csv(orig_stats_p2_pwr_file,
                          orig_stats_p2_pwr_moddate)),
  
  # orig_prereg-indicated.csv ----
  # Indicates whether the original study paper mentions preregistration
  tar_target(orig_prereg_indicated_file,
             "2k34q"),
  tar_target(orig_prereg_indicated_moddate,
             get_osf_mod_date(orig_prereg_indicated_file),
             cue = tar_cue("always")),
  tar_target(orig_prereg_indicated,
             load_osf_csv(orig_prereg_indicated_file,
                          orig_prereg_indicated_moddate,
                          col_types = c("clcd")))
)
```

### Replication Studies

```{targets raw-repli}
list(
  # repli_effective.csv ----
  # Adjusted sample sizes and degrees of freedom for the purpose of 
  # effect size estimation.
  tar_target(repli_effective_file,
             "mhf8d"),
  tar_target(repli_effective_moddate,
             get_osf_mod_date(repli_effective_file),
             cue = tar_cue("always")),
  tar_target(repli_effective,
             load_osf_csv(repli_effective_file,
                          repli_effective_moddate)),

  # repli_effect-size.csv ----
  # Calculated effect sizes with upper and lower bounds for replication studies
  tar_target(repli_effect_size_file,
             "yxw5t"),
  tar_target(repli_effect_size_moddate,
             get_osf_mod_date(repli_effect_size_file),
             cue = tar_cue("always")),
  tar_target(repli_effect_size,
             load_osf_csv(repli_effect_size_file,
                          repli_effect_size_moddate)),
  
  # repli_extended.csv ----
  # Additional statistics for replication studies calculated by collaborators 
  # at Tilburg University.
  tar_target(repli_extended_file,
             "e9v5x"),
  tar_target(repli_extended_moddate, 
             get_osf_mod_date(repli_extended_file),
             cue = tar_cue("always")), 
  tar_target(repli_extended,
             load_osf_csv(repli_extended_file,
                          repli_extended_moddate)),
  
  # repli_stats_p1.csv ----
  # Statistics from replication studies in phase 1.
  tar_target(repli_stats_p1_file,
             "cyr63"),
  tar_target(repli_stats_p1_moddate,
             get_osf_mod_date(repli_stats_p1_file),
             cue = tar_cue("always")),
  tar_target(repli_stats_p1,
             load_osf_csv(repli_stats_p1_file,
                          repli_stats_p1_moddate)),
  
  # repli_stats_p2_additional.csv ----
  # Statistics from replication studies in phase 2 that were entered 
  # manually later in the process
  tar_target(repli_stats_p2_additional_file,
             "6jhy7"),
  tar_target(repli_stats_p2_additional_moddate,
             get_osf_mod_date(repli_stats_p2_additional_file),
             cue = tar_cue("always")),
  tar_target(repli_stats_p2_additional,
             load_osf_csv(repli_stats_p2_additional_file,
                          repli_stats_p2_additional_moddate)),
  
  # repli_stats_p2.csv ----
  # Statistics from replication studies in phase 2 that were entered via 
  # google form.
  tar_target(repli_stats_p2_file,
             "7xfpq"),
  tar_target(repli_stats_p2_moddate,
             get_osf_mod_date(repli_stats_p2_file),
             cue = tar_cue("always")),
  tar_target(repli_stats_p2,
             load_osf_csv(repli_stats_p2_file,
                          repli_stats_p2_moddate))
)
```

### Reproduction Studies

```{targets raw-repro}
list(
  # repro_prereg-check.csv ----
  # Researcher check-in information for reproduction registrations
  tar_target(repro_prereg_check_file,
             "7kydz"),
  tar_target(repro_prereg_check_moddate,
             get_osf_mod_date(repro_prereg_check_file),
             cue = tar_cue("always")),
  tar_target(repro_prereg_check,
             load_osf_csv(repro_prereg_check_file,
                               repro_prereg_check_moddate)),
  
  # repro_stats_additional.csv ----
  # Statistics from reproduction studies that were entered later in the 
  # process rather than through the google form
  tar_target(repro_stats_additional_file,
             "sw49d"),
  tar_target(repro_stats_additional_moddate,
             get_osf_mod_date(repro_stats_additional_file),
             cue = tar_cue("always")),
  tar_target(repro_stats_additional,
             load_osf_csv(repro_stats_additional_file,
                          repro_stats_additional_moddate)),
  
  # repro_stats.csv ----
  # Statistics from reproduction studies that were entered via google form.
  tar_target(repro_stats_file,
             "h82yc"),
  tar_target(repro_stats_moddate,
             get_osf_mod_date(repro_stats_file),
             cue = tar_cue("always")),
  tar_target(repro_stats,
             load_osf_csv(repro_stats_file,
                          repro_stats_moddate)),
  
  # repro_vor.csv ----
  # Source of the reproduction version of record in cases with multiple
  # reproductions for the same claim
  tar_target(repro_vor_file,
             "k8e7w"),
  tar_target(repro_vor_moddate,
             get_osf_mod_date(repro_vor_file),
             cue = tar_cue("always")),
  tar_target(repro_vor,
             load_osf_csv(repro_vor_file,
                          repro_vor_moddate))
)
```

### Process Reproducibility

```{targets raw-pr}
list(
  # orig-material_author-outreach.csv ----
  # Original study data and code availability based on author outreach efforts
  tar_target(materials_author_outreach_file,
             "uxj48"),
  tar_target(materials_author_outreach_moddate,
             get_osf_mod_date(materials_author_outreach_file),
             cue = tar_cue("always")),
  tar_target(materials_author_outreach,
             load_osf_csv(materials_author_outreach_file,
                          materials_author_outreach_moddate)),
  
  # orig-material_availability.csv ----
  # Original study data and code availability determinations from a 
  # variety of sources
  tar_target(materials_availability_file,
             "dz5cs"),
  tar_target(materials_availability_moddate,
             get_osf_mod_date(materials_availability_file),
             cue = tar_cue("always")),
  tar_target(materials_availability,
             load_osf_csv(materials_availability_file,
                          materials_availability_moddate))
)
```

### Project Management

Some information comes from datasets used in project management.

```{targets raw-project}
list(
  # score_paper_id-key.csv ----
  # Links canonical paper ID to the IDs used by other teams. 
  # Does not include COVID papers
  tar_target(score_paper_id_key_file,
             "qgdwh"),
  tar_target(score_paper_id_key_moddate,
             get_osf_mod_date(score_paper_id_key_file),
             cue = tar_cue("always")),
  tar_target(score_paper_id_key,
             load_osf_csv(score_paper_id_key_file,
                          score_paper_id_key_moddate)),
  
  # score_paper_status.csv ----
  # Status of articles at each stage in SCORE. 
  # Does not contain COVID preprints.
  tar_target(score_paper_status_file,
             "smpd6"),
  tar_target(score_paper_status_moddate,
             get_osf_mod_date(score_paper_status_file),
             cue = tar_cue("always")),
  tar_target(score_paper_status,
             load_osf_csv(score_paper_status_file,
                          score_paper_status_moddate)),
  
  # score_rr-projects_dates.csv ----
  # Start and end dates of replication/reproduction projects
  tar_target(score_rr_dates_file,
             "6tpf7"),
  tar_target(score_rr_dates_moddate,
             get_osf_mod_date(score_rr_dates_file),
             cue = tar_cue("always")),
  tar_target(score_rr_dates,
             load_osf_csv(score_rr_dates_file,
                          score_rr_dates_moddate,
                          col_types = c("cccDDD"))),
  
  # score_rr-projects_not-sourced.csv ----
  # Reasons why non-matched papers were not plausible for replication
  tar_target(score_rr_not_sourced_file,
             "3hxzj"),
  tar_target(score_rr_not_sourced_moddate,
             get_osf_mod_date(score_rr_not_sourced_file),
             cue = tar_cue("always")),
  tar_target(score_rr_not_sourced,
             load_osf_csv(score_rr_not_sourced_file,
                          score_rr_not_sourced_moddate)),
  
  # score_rr-projects_osf-ids_add.csv ----
  # Links to replication and reproduction OSF projects and components for
  # additional projects not in the original tracking file.
  tar_target(score_rr_osf_ids_add_file,
             "et6ng"),
  tar_target(score_rr_osf_ids_add_moddate,
             get_osf_mod_date(score_rr_osf_ids_add_file),
             cue = tar_cue("always")),
  tar_target(score_rr_osf_ids_add,
             load_osf_csv(score_rr_osf_ids_add_file,
                          score_rr_osf_ids_add_moddate)),
  
  # score_rr-projects_osf-ids_raw.csv ----
  # Links to replication and reproduction OSF projects and components.
  tar_target(score_rr_osf_ids_raw_file,
             "7nmgy"),
  tar_target(score_rr_osf_ids_raw_moddate,
             get_osf_mod_date(score_rr_osf_ids_raw_file),
             cue = tar_cue("always")),
  tar_target(score_rr_osf_ids_raw,
             load_osf_csv(score_rr_osf_ids_raw_file,
                          score_rr_osf_ids_raw_moddate)),
  
  # score_rr-projects_sourced.csv ----
  # All replication and reproduction projects that were matched with a team. 
  tar_target(score_rr_sourced_file,
             "x2cbe"),
  tar_target(score_rr_sourced_moddate,
             get_osf_mod_date(score_rr_sourced_file),
             cue = tar_cue("always")),
  tar_target(score_rr_sourced,
             load_osf_csv(score_rr_sourced_file,
                          score_rr_sourced_moddate)),

  # score_rr-project_status.csv ----
  # Status tracking and other internal documentation for replication and
  # reproduction studies
  tar_target(score_rr_status_file,
             "khpbu"),
  tar_target(score_rr_status_moddate,
             get_osf_mod_date(score_rr_status_file),
             cue = tar_cue("always")),
  tar_target(score_rr_status,
             load_osf_csv(score_rr_status_file,
                          score_rr_status_moddate)),
  
  # score_rr-project_types.csv ----
  # Internal project types for replication and reproduction studies. 
  tar_target(score_rr_types_file,
             "dx759"),
  tar_target(score_rr_types_moddate,
             get_osf_mod_date(score_rr_types_file),
             cue = tar_cue("always")),
  tar_target(score_rr_types_raw,
             load_osf_csv(score_rr_types_file,
                          score_rr_types_moddate))
)
```

### Correlations

```{targets raw-corr}
list(
  # corr_orig-significance_p1.csv ----
  # Original study claims significance for phase 1 claims
  tar_target(stat_sig_p1_file,
             "ftxyn"),
  tar_target(stat_sig_p1_moddate,
             get_osf_mod_date(stat_sig_p1_file),
             cue = tar_cue("always")),
  tar_target(stat_sig_p1,
             load_osf_csv(stat_sig_p1_file,
                          stat_sig_p1_moddate)),
  
  # corr_orig-significance_p2.csv ----
  # Original study claims significance for phase 2 claims
  tar_target(stat_sig_p2_file,
             "kqp2w"),
  tar_target(stat_sig_p2_moddate,
             get_osf_mod_date(stat_sig_p2_file),
             cue = tar_cue("always")),
  tar_target(stat_sig_p2,
             load_osf_csv(stat_sig_p2_file,
                          stat_sig_p2_moddate))
)
```


## Change Logs

If changes ever needed to be made to any dataset for any reason, these changes were recorded in a change log. Change logs record who requested the change, when the change was implemented, the key ID of the record to change, the column the change was in, what the change was to, what data was recorded there previously, and the reasoning behind the change.

```{targets change-logs}
list(
  # changes_journal_metadata.csv ----
  # Change log for journal metadata
  tar_target(changes_journal_metadata_file,
             "c2vdn"),
  tar_target(changes_journal_metadata_moddate,
             get_osf_mod_date(changes_journal_metadata_file),
             cue = tar_cue("always")),
  tar_target(changes_journal_metadata,
             load_osf_csv(changes_journal_metadata_file,
                          changes_journal_metadata_moddate)),
  
  # changes_orig_merged.csv ----
  # Change log for merged original study variables
  tar_target(changes_orig_merged_file,
             "4azsd"),
  tar_target(changes_orig_merged_moddate,
             get_osf_mod_date(changes_orig_merged_file),
             cue = tar_cue("always")),
  tar_target(changes_orig_merged,
             load_osf_csv(changes_orig_merged_file,
                          changes_orig_merged_moddate)),
  
  # changes_orig_stats-extended.csv ----
  # Change log for additional statistics for original studies
  tar_target(changes_orig_stats_extended_file,
             "vz2ck"),
  tar_target(changes_orig_stats_extended_moddate,
             get_osf_mod_date(changes_orig_stats_extended_file),
             cue = tar_cue("always")),
  tar_target(changes_orig_stats_extended,
             load_osf_csv(changes_orig_stats_extended_file,
                          changes_orig_stats_extended_moddate)),
  
  # changes_pr_data-form.csv ----
  # Change log for process reproducibility variables
  tar_target(changes_pr_data_form_file,
             "pcyjg"),
  tar_target(changes_pr_data_form_moddate,
             get_osf_mod_date(changes_pr_data_form_file),
             cue = tar_cue("always")),
  tar_target(changes_pr_data_form,
             load_osf_csv(changes_pr_data_form_file,
                          changes_pr_data_form_moddate)),
  
  # changes_pr_outreach.csv ----
  # Change log for original study author outreach variables
  tar_target(changes_author_outreach_file,
             "bhpuv"),
  tar_target(changes_author_outreach_moddate,
             get_osf_mod_date(changes_author_outreach_file),
             cue = tar_cue("always")),
  tar_target(changes_author_outreach,
             load_osf_csv(changes_author_outreach_file,
                          changes_author_outreach_moddate)),
  
  # changes_repli_merged.csv ----
  # Change log for merged replication study variables
  tar_target(changes_repli_merged_file,
             "uq7pd"),
  tar_target(changes_repli_merged_moddate,
             get_osf_mod_date(changes_repli_merged_file),
             cue = tar_cue("always")),
  tar_target(changes_repli_merged,
             load_osf_csv(changes_repli_merged_file,
                          changes_repli_merged_moddate)),
  
  # changes_repro_merged.csv ----
  # Change log for merged reproduction study variables
  tar_target(changes_repro_merged_file,
             "hs4v7"),
  tar_target(changes_repro_merged_moddate,
             get_osf_mod_date(changes_repro_merged_file),
             cue = tar_cue("always")),
  tar_target(changes_repro_merged,
             load_osf_csv(changes_repro_merged_file,
                          changes_repro_merged_moddate)),
  
  # changes_repro_summary.csv ----
  # Change log for reproduction summary variables
  tar_target(changes_repro_summary_file,
             "e498m"),
  tar_target(changes_repro_summary_moddate,
             get_osf_mod_date(changes_repro_summary_file),
             cue = tar_cue("always")),
  tar_target(changes_repro_summary,
             load_osf_csv(changes_repro_summary_file,
                          changes_repro_summary_moddate)),
  
  # changes_score_rr_status.csv ----
  # Change log for score replication/reproduction project status
  tar_target(changes_score_rr_status_file,
             "9wfc6"),
  tar_target(changes_score_rr_status_moddate,
             get_osf_mod_date(changes_score_rr_status_file),
             cue = tar_cue("always")),
  tar_target(changes_score_rr_status,
             load_osf_csv(changes_score_rr_status_file,
                          changes_score_rr_status_moddate)),
  
  # changes_score_rr_types.csv ----
  # Change log for score internal project types
  tar_target(changes_score_rr_types_file,
             "xdny8"),
  tar_target(changes_score_rr_types_moddate,
             get_osf_mod_date(changes_score_rr_types_file),
             cue = tar_cue("always")),
  tar_target(changes_score_rr_types,
             load_osf_csv(changes_score_rr_types_file,
                          changes_score_rr_types_moddate))
)
```

## Intermediate Data

These are data at various stages of processing between the raw data and analysis-ready data. 

### Project Management

```{targets int-project}
list(
  # Updated replication/reproduction project types
  tar_target(score_rr_types,
             update_score_rr_types(score_rr_types_raw,
                                   changes_score_rr_types))
)
```

### Journal Metadata

```{targets int-journal}
list(
  # Update and transform journal metadata
  tar_target(journal_updated,
             transform_update_journal(journal_metadata_raw,
                                      changes_journal_metadata)),
  
  # Join journal metadata and policy information
  tar_target(journal_metadata,
             left_join(journal_updated,
                       journal_top_factor,
                       by = join_by(publication_standard == Journal)))
)
```

### Paper Metadata

```{targets int-paper}
list(
  # Metadata for covid papers was collected separately from other papers
  # Reconcile differences for merging
  tar_target(papers_covid_transformed,
             transform_covid_metadata(paper_metadata_covid)),
  
  tar_target(papers_transformed,
             transform_paper_metadata(paper_metadata_raw, 
                                      score_paper_status))
)
```

### Original Studies

```{targets int-orig}
list(
  # Transformed and merged raw original statistics data
  tar_target(orig_merged,
             merge_orig_stats(orig_stats_p1,
                              orig_claims_p1_covid,
                              orig_stats_p2_pwr,
                              orig_stats_p2_manual,
                              orig_stats_p2_form,
                              orig_stats_p2_additional)),
  
  # Merged original statistics updated with changes
  tar_target(orig_updated,
             update_orig_stats(orig_merged,
                               changes_orig_merged)),
  
  # Extended original study statistics updated with changes
  tar_target(orig_extended_updated,
             update_orig_extended(orig_stats_extended,
                                  changes_orig_stats_extended)),
  
  # Original study statistics, merged and transformed into a semi-final form
  tar_target(orig_dataset,
             transform_orig_updated(orig_updated,
                                    orig_extended_updated))
)
```

### Replication Studies

```{targets int-repli}
list(
  # Transformed replication form data
  tar_target(repli_stats_p2_transformed,
             transform_repli_stats_p2(repli_stats_p2, 
                                      score_rr_types)),
  
  # Merged replication study variables
  tar_target(repli_merged,
             merge_repli_input(repli_stats_p1,
                               repli_stats_p2_transformed,
                               repli_stats_p2_additional,
                               repli_effective)),
  
  # Replication study statistics, updated and transformed 
  # into a semi-final form
  tar_target(repli_dataset,
             update_transform_repli(repli_merged, 
                                    changes_repli_merged))
)
```

### Reproduction Studies

```{targets int-repro}
list(
  # Transformed data from forms
  tar_target(repro_data_entry,
             transform_repro_input(repro_stats)),
  
  # Transformed reproduction preregistration check-ins
  tar_target(rr_confrontations_repro_claim,
             transform_repro_prereg(repro_prereg_check)),
  
  # Merged reproduction study statistics
  tar_target(repro_merged,
             merge_repro(repro_data_entry,
                         repro_stats_additional,
                         repro_vor)),
  
  # Updated and transformed reproduction study statistics
  tar_target(repro_dataset,
             update_repro(repro_merged,
                          changes_repro_merged)),
  
  # Reproduction study summary variables
  tar_target(repro_summary_raw,
             create_repro_summary(repro_dataset,
                                  orig_dataset,
                                  rr_confrontations_repro_claim)),
  
  # Reproduction study summary variables updated with new categories
  tar_target(repro_summary,
             update_repro_summary(repro_summary_raw,
                                  changes_repro_summary))
)
```


### Process Reproducibility

```{targets int-pr}
list(
  # Updated original author outreach 
  tar_target(oa_outreach,
             update_oa(materials_author_outreach,
                       changes_author_outreach)),
  
  # Transformed process reproducibility 
  tar_target(pr_transformed,
             transform_pr(materials_availability,
                          orig_claims_covid_ta2,
                          orig_claims_p1_covid)),
  
  # Updated process reproducibility 
  tar_target(pr_updated,
             update_pr(pr_transformed,
                       changes_pr_data_form))
  
)
```

## Analysis-ready Data

Datasets that were used in papers produced during the SCORE project.

### Replication and Reproduction Papers

```{targets analytic-repli-repro}
list(
  # claims_bushel_ids.csv ----
  # Key IDs for bushel claims, where multiple inferential tests were evaluated 
  # within a single study
  # https://osf.io/dtzx4/files/kxr7p
  tar_target(claims_bushel_ids,
             select(orig_claims_p2_bushel,
                    paper_id,
                    claim4_id)),
  
  # claims_non_signigicant_bushels.csv ----
  # Which bushel claims use non-significant evidence
  # https://osf.io/dtzx4/files/4az3m
  tar_target(claims_non_significant_bushels,
             select(orig_claims_bushel_sig,
                    paper_id,
                    claim4_id,
                    coded_stat_evidence,
                    nonsig)),
  
  # claims_ser_power.csv ----
  # Set of sample sizes calculated via the SER method
  # https://osf.io/dtzx4/files/xscnj
  tar_target(claims_ser_power,
             select(orig_n_ser,
                    claim_id = unique_claim_id,
                    rr_sample_size_50_for_100 = threshold,
                    rr_sample_size_90_for_75 = s1,
                    rr_sample_size_90_for_50 = s2)),
  
  # claims_traditional_power.csv ----
  # Set of sample sizes calculated via traditional methods
  # https://osf.io/dtzx4/files/dkqns
  tar_target(claims_traditional_power,
             select(
               orig_n_traditional,
               claim_id,
               rr_sample_size_50_for_100 = rr_threshold_analytic_sample_size,
               rr_sample_size_90_for_75 = rr_stage1_analytic_sample_size,
               rr_sample_size_90_for_50 = rr_stage2_analytic_sample_size
              )),
  
  # extracted_claims.csv ----
  # Claims extracted from original studies
  # https://osf.io/dtzx4/files/j2fp5
  tar_target(extracted_claims,
             merge_claims(orig_claims_p1,
                          orig_claims_p1_covid,
                          orig_claims_p2_single,
                          orig_claims_p2_bushel,
                          orig_claims_bushel_complex,
                          score_paper_status,
                          papers_covid_transformed)),
  
  # llm_method_data.csv ----
  # Summarizes the prevalence and scoring of various  
  # statistical/methodological categories across 173 research abstracts.
  tar_target(llm_method_data_file,
             "vxysn"),
  tar_target(llm_method_data_moddate,
             get_osf_mod_date(llm_method_data_file)),
  tar_target(llm_method_data,
             load_osf_csv(llm_method_data_file,
                          llm_method_data_moddate)),
  
  # llm_research_question_table.csv ----
  # This dataset summarizes the research questions across 173 
  # research abstracts.
  tar_target(llm_research_question_table_file,
             "mg794"),
  tar_target(llm_research_question_table_moddate,
             get_osf_mod_date(llm_research_question_table_file)),
  tar_target(llm_research_question_table,
             load_osf_csv(llm_research_question_table_file,
                          llm_research_question_table_moddate)),
  
  # llm_theory_data.csv ----
  # Summarizes the prevalence and scoring of various theoretical paradigms 
  # across 173 research abstracts.
  tar_target(llm_theory_data_file,
             "mp3rq"),
  tar_target(llm_theory_data_moddate,
             get_osf_mod_date(llm_theory_data_file)),
  tar_target(llm_theory_data,
             load_osf_csv(llm_theory_data_file,
                          llm_theory_data_moddate)),
  
  # orig_outcomes.csv ----
  # Statistics from original studies from which claims were extracted
  # https://osf.io/dtzx4/files/y7bq5
  tar_target(orig_outcomes,
             create_ov_analytic(orig_dataset,
                                orig_claims_bushel_complex,
                                orig_stats_effect,
                                orig_effect_size,
                                repli_dataset)),
  
  # paper_metadata.csv ----
  # Metadata for original study articles
  # https://osf.io/dtzx4/files/jg54h
  tar_target(paper_metadata,
             merge_paper_metadata(papers_transformed,
                                  papers_covid_transformed,
                                  paper_citations,
                                  journal_metadata)),

  # papers_not_sourced.csv ----
  # Accounting for all papers that were never sourced for a replication 
  # attempt
  # https://osf.io/dtzx4/files/rdzc5
  tar_target(papers_not_sourced,
             select(score_rr_not_sourced,
                    paper_id,
                    dr_plausible,
                    dr_reason,
                    dar_plausible,
                    dar_reason)),
  
  # paper_status_tracking.csv ----
  # An internal tracking dataset showing which stages of the project each 
  # paper was involved in
  # https://osf.io/dtzx4/files/yag6n
  tar_target(paper_status_tracking,
             select(score_paper_status,
                    paper_id,
                    prepool,
                    pool,
                    presample,
                    sample,
                    rejected,
                    RR,
                    p1_delivery,
                    p2_presample,
                    p2_sample,
                    p2_delivery,
                    bushel)),
  
  # pr_outcomes.csv ----
  # "Process reproducibility", or assessing whether the findings within an 
  # original paper could theoretically be reproduced using the original data 
  # and code to reproduce the results reported in the original paper, as well 
  # as how feasible conducting a reproduction might be, based on the 
  # availability of original materials necessary to reproduce the findings of 
  # a given paper.
  # https://osf.io/dtzx4/files/e58n9
  tar_target(pr_outcomes,
             create_pr_analytic(pr_updated, oa_outreach)),
  
  # repli_all_cases.csv ----
  # Replication outcomes, including cases that were excluded from analysis for 
  # non-significant findings, incomplete/insufficient data collection, or 
  # power/sample size issues.
  # https://osf.io/dtzx4/files/2k9ht
  tar_target(repli_all_cases,
             select(repli_dataset,
                    rr_id,
                    paper_id,
                    rr_type_internal,
                    report_id = unique_report_id,
                    rr_repl_exact_replicated_reference)),
  
  # repli_binary.csv ----
  # A series of boolean fields describing replication measures
  # https://osf.io/dtzx4/files/q6rtd
  tar_target(repli_binary,
             calculate_repli_binary(repli_outcomes,
                                    orig_outcomes,
                                    paper_metadata)),
  
  # repli_exclusions.csv ----
  # The replication findings excluded from the main replication reporting 
  # and their exclusion reasons
  tar_target(repli_case_exclusions_file,
             "wq65j"),
  tar_target(repli_case_exclusions_moddate,
             get_osf_mod_date(repli_case_exclusions_file),
             cue = tar_cue("always")), 
  tar_target(repli_case_exclusions,
             load_osf_csv(repli_case_exclusions_file,
                          repli_case_exclusions_moddate)),
  
  # repli_outcomes.csv ----
  # Replications of the original studies from which claims were extracted
  # https://osf.io/dtzx4/files/2c8gh
  tar_target(repli_outcomes,
             create_repli_analytic(repli_dataset,
                                   score_rr_types,
                                   repli_extended,
                                   score_paper_status,
                                   orig_claims_p2_bushel,
                                   repli_effect_size,
                                   orig_outcomes,
                                   paper_metadata,
                                   repli_case_exclusions)),
  
  # repro_journal_policies.csv ----
  # Journal policies about reproducibility and data/code sharing
  # Renamed to keep consistent with analysis code
  # https://osf.io/dtzx4/files/6t8gy
  tar_target(repro_journal_policies,
             journal_repro_policies),
  
  # repro_outcomes.csv ----
  # Reproductions of the original studies from which claims were extracted
  # https://osf.io/dtzx4/files/47jhm
  tar_target(repro_outcomes,
             create_repro_analytic(repro_dataset,
                                   repro_summary)),
  
  # rr_internal_tracking.csv ----
  # This is a convenience dataset used by COS staff for a variety of purposes. 
  # Some variables, like outcome and results_available, reflect the particular 
  # time that the variables were created and are no longer meaningful 
  # reflections of the present state of the SCORE data. For that reason, 
  # external use of these variables should be done cautiously. 
  # https://osf.io/dtzx4/files/2e5nk
  tar_target(rr_internal_tracking,
             score_rr_status %>%
               add_column(status_version = 1) %>%
               apply_changelog(changes_score_rr_status,
                               "rr_id") %>%
               select(rr_id,
                    paper_id,
                    project_guid,
                    field,
                    type,
                    outcome,
                    results_available,
                    registrations,
                    prereg_completion,
                    osf_activity,
                    all_types)),
  
  # rr_project_dates.csv ----
  # Milestone dates of replication studies
  # https://osf.io/dtzx4/files/n629a
  tar_target(rr_project_dates,
             select(score_rr_dates,
                    rr_id,
                    project_guid,
                    type,
                    started,
                    reg_date,
                    completed)),

  # rr_type_key.csv ----
  # Dataset of replication and reproduction projects sourced to a research  
  # team for completion, including their project types.
  # https://osf.io/dtzx4/files/abq68
  tar_target(rr_type_key,
             select(score_rr_sourced,
                    rr_id,
                    paper_id,
                    type))
)
```

### Correlations Paper Datasets

```{targets analytic-correlations}
list(
  # corr_citation-counts.csv ----
  # Original author citation counts
  tar_target(oa_citations_file,
             "bsfwk"),
  tar_target(oa_citations_moddate,
             get_osf_mod_date(oa_citations_file),
             cue = tar_cue("always")),
  tar_target(oa_citations,
             load_osf_csv(oa_citations_file,
                          oa_citations_moddate)),
  
  # corr_dataseer_orig-metadata.csv ----
  # Extended metadata and data/code availability from DataSeer
  tar_target(dataseer_file,
             "cdvh2"),
  tar_target(dataseer_moddate,
             get_osf_mod_date(dataseer_file),
             cue = tar_cue("always")),
  tar_target(dataseer,
             load_osf_csv(dataseer_file,
                          dataseer_moddate)),
  
  # corr_ksu_bibliometric.csv ----
  # Automated bibliometric analyses from KSU
  tar_target(ksu_file,
             "uenaj"),
  tar_target(ksu_moddate,
             get_osf_mod_date(ksu_file),
             cue = tar_cue("always")),
  tar_target(ksu,
             load_osf_csv(ksu_file,
                          ksu_moddate)),
  
  # corr_multi100.csv ----
  # Multi100 Robustness dataset
  tar_target(m100_file,
             "v2r4g"),
  tar_target(m100_moddate,
             get_osf_mod_date(m100_file),
             cue = tar_cue("always")),
  tar_target(m100,
             load_osf_csv(m100_file,
                          m100_moddate)),
  
  # corr_orig-effect-size.csv ----
  # Original study claims effect sizes and their source
  tar_target(es_file,
             "7sc4f"),
  tar_target(es_moddate,
             get_osf_mod_date(es_file),
             cue = tar_cue("always")),
  tar_target(es,
             load_osf_csv(es_file,
                          es_moddate)),
  
  # corr_orig-significance.csv ----
  # Original study claims significance
  # https://osf.io/dtzx4/files/d5qke
  tar_target(sig,
             rbind(stat_sig_p1, stat_sig_p2)),
  
  # corr_scholarcy.csv ----
  # Extracted open science indicators from scholarcy
  tar_target(scholarcy_file,
             "68xp3"),
  tar_target(scholarcy_moddate,
             get_osf_mod_date(scholarcy_file),
             cue = tar_cue("always")),
  tar_target(scholarcy,
             load_osf_csv(scholarcy_file,
                          scholarcy_moddate)),
  
  # corr_sciscore.csv ----
  # SciScore report, scoring SCORE papers on a ten point rigor and 
  # transparency index. 
  # Delivered to The Center For Open Science from SciCrunch.
  tar_target(sciscore_file,
             "7u2pa"),
  tar_target(sciscore_moddate,
             get_osf_mod_date(sciscore_file),
             cue = tar_cue("always")),
  tar_target(sciscore,
             load_osf_csv(sciscore_file,
                          sciscore_moddate)),
  
  # corr_sit_orig-authors.csv ----
  # Original author metadata from the Stevens Institute of Technology
  tar_target(oa_metadata_file,
             "38upf"),
  tar_target(oa_metadata_moddate,
             get_osf_mod_date(oa_metadata_file),
             cue = tar_cue("always")),
  tar_target(oa_metadata,
             load_osf_csv(oa_metadata_file,
                          oa_metadata_moddate)),
  
  # corr_statcheck.csv ----
  # Statcheck report of Null Hypothesis Significance test (NHST) results from
  # SCORE papers assessing whether the extracted results are 
  # internally consistent.
  tar_target(statcheck_file,
             "jm2ka"),
  tar_target(statcheck_moddate,
             get_osf_mod_date(statcheck_file),
             cue = tar_cue("always")),
  tar_target(statcheck,
             load_osf_csv(statcheck_file,
                          statcheck_moddate)),

  # corr_usc_hedging-humility.csv ----
  # Hedging and humility data from the University of Southern California
  tar_target(hedge_file,
             "tj36g"),
  tar_target(hedge_moddate,
             get_osf_mod_date(hedge_file),
             cue = tar_cue("always")),
  tar_target(hedge,
             load_osf_csv(hedge_file,
                          hedge_moddate))
)
```

#### Confidence Scores

```{targets corr-confidence-scores}
list(
  # Prediction Markets ----
  ## corr_keywest_cs.csv ----
  ## Prediction markets confidence scores
  tar_target(kw_scores_file,
             "gdw29"),
  tar_target(kw_scores_moddate, 
             get_osf_mod_date(kw_scores_file),
             cue = tar_cue("always")), 
  tar_target(kw_scores,
             load_osf_csv(kw_scores_file,
                          kw_scores_moddate)),

  # Synthetic Markets ----
  ## corr_psu_cs_1.csv ----
  ## Sythentic markets confidence scores for sprint 1
  tar_target(psu_s1_file,
             "ch9dk"),
  tar_target(psu_s1_moddate, 
             get_osf_mod_date(psu_s1_file),
             cue = tar_cue("always")), 
  tar_target(psu_s1,
             load_osf_csv(psu_s1_file,
                          psu_s1_moddate)),
  
  ## corr_psu_cs_2.csv ----
  ## Sythentic markets confidence scores for sprint 2
  tar_target(psu_s2_file,
             "ayjdb"),
  tar_target(psu_s2_moddate, 
             get_osf_mod_date(psu_s2_file),
             cue = tar_cue("always")), 
  tar_target(psu_s2,
             load_osf_csv(psu_s2_file,
                          psu_s2_moddate)),
  
  ## corr_psu_cs_3.csv ----
  ## Sythentic markets confidence scores for sprint 3
  tar_target(psu_s3_file,
             "tkqax"),
  tar_target(psu_s3_moddate, 
             get_osf_mod_date(psu_s3_file),
             cue = tar_cue("always")), 
  tar_target(psu_s3,
             load_osf_csv(psu_s3_file,
                          psu_s3_moddate)),
  
  # A+ ----
  ## corr_twosix_cs_1.csv ----
  ## A+ confidence scores for sprint 1
  tar_target(ts_s1_file,
             "r245a"),
  tar_target(ts_s1_moddate, 
             get_osf_mod_date(ts_s1_file),
             cue = tar_cue("always")), 
  tar_target(ts_s1,
             load_osf_csv(ts_s1_file,
                          ts_s1_moddate)),
  
  ## corr_twosix_cs_2.csv ----
  ## A+ confidence scores for sprint 2
  tar_target(ts_s2_file,
             "fzchp"),
  tar_target(ts_s2_moddate, 
             get_osf_mod_date(ts_s2_file),
             cue = tar_cue("always")), 
  tar_target(ts_s2,
             load_osf_csv(ts_s2_file,
                          ts_s2_moddate)),
  
  ## corr_twosix_cs_3.csv ----
  ## A+ confidence scores for sprint 3
  tar_target(ts_s3_file,
             "tje32"),
  tar_target(ts_s3_moddate, 
             get_osf_mod_date(ts_s3_file),
             cue = tar_cue("always")), 
  tar_target(ts_s3,
             load_osf_csv(ts_s3_file,
                          ts_s3_moddate)),
  
  ## corr_twosix_cs_bushel.csv ----
  ## A+ confidence scores for bushel claims
  tar_target(ts_bushel_file,
             "fz3w9"),
  tar_target(ts_bushel_moddate, 
             get_osf_mod_date(ts_bushel_file),
             cue = tar_cue("always")), 
  tar_target(ts_bushel,
             load_osf_csv(ts_bushel_file,
                          ts_bushel_moddate)),
  
  # Elicitations ----
  ## corr_umelb_cs_1.csv ----
  ## Elicitations team confidence scores for phase 1
  tar_target(melb_p1_file,
             "p7x53"),
  tar_target(melb_p1_moddate, 
             get_osf_mod_date(melb_p1_file),
             cue = tar_cue("always")), 
  tar_target(melb_p1,
             load_osf_csv(melb_p1_file,
                          melb_p1_moddate)),
  
  ## corr_umelb_cs_2.csv ----
  ## Elicitations team confidence scores for phase 2
  tar_target(melb_p2_file,
             "p8t4d"),
  tar_target(melb_p2_moddate, 
             get_osf_mod_date(melb_p2_file),
             cue = tar_cue("always")), 
  tar_target(melb_p2,
             load_osf_csv(melb_p2_file,
                          melb_p2_moddate)),
  
  ## corr_umelb_cs_bushel.csv ----
  ## Elicitations team confidence scores for bushel claims
  tar_target(melb_bushel_file,
             "5mr6c"),
  tar_target(melb_bushel_moddate, 
             get_osf_mod_date(melb_bushel_file),
             cue = tar_cue("always")), 
  tar_target(melb_bushel,
             load_osf_csv(melb_bushel_file,
                          melb_bushel_moddate)),
  
  # MACROSCORE ----
  ## corr_usc_cs_1.csv ----
  ## MACROSCORE confidence scores for sprint 1
  tar_target(usc_s1_file,
             "uhavw"),
  tar_target(usc_s1_moddate, 
             get_osf_mod_date(usc_s1_file),
             cue = tar_cue("always")), 
  tar_target(usc_s1,
             load_osf_csv(usc_s1_file,
                          usc_s1_moddate)),
  
  ## corr_usc_cs_2.csv ----
  ## MACROSCORE confidence scores for sprint 2
  tar_target(usc_s2_file,
             "djtnh"),
  tar_target(usc_s2_moddate, 
             get_osf_mod_date(usc_s2_file),
             cue = tar_cue("always")), 
  tar_target(usc_s2,
             load_osf_csv(usc_s2_file,
                          usc_s2_moddate)),
  
  ## corr_usc_cs_3.csv ----
  ## MACROSCORE confidence scores for sprint 3
  tar_target(usc_s3_file,
             "mf74q"),
  tar_target(usc_s3_moddate, 
             get_osf_mod_date(usc_s3_file),
             cue = tar_cue("always")), 
  tar_target(usc_s3,
             load_osf_csv(usc_s3_file,
                          usc_s3_moddate)),
  
  ## corr_usc_cs_bushel.csv ----
  ## MACROSCORE confidence scores for bushel claims
  tar_target(usc_bushel_file,
             "ta4sh"),
  tar_target(usc_bushel_moddate, 
             get_osf_mod_date(usc_bushel_file),
             cue = tar_cue("always")), 
  tar_target(usc_bushel,
             load_osf_csv(usc_bushel_file,
                          usc_bushel_moddate))
)
```

### Other

```{targets analytic-other}
list(
  # A list of all reproduction/replication projects with their OSF GUIDs
  tar_target(score_rr_osf_ids,
             extend_rr_projects(score_rr_osf_ids_raw,
                                score_rr_osf_ids_add))
)
```


# Run Pipeline

Run the `{targets}` pipeline. Expected warnings:

- repli_stats_p2_transformed: There were 2 warnings in `mutate()`.
- repli_binary: There was 1 warning in `mutate()`.


```{r make}
tar_make()

targets::tar_meta(fields = warnings, complete_only = TRUE)
```

# Output

You can retrieve results from the `_targets/` data store using `tar_read()` or `tar_load()`.

The `targets` dependency graph shows the steps of the pipeline at a high level.

```{r visnetwork}
tar_visnetwork()
```