{
 "title": "North Korean industrial enterprises in state media, 2012-2025",
 "version": "2.0 (workshop draft, 2026-10-30)",
 "supersedes": "1.0 (eight sectors, seven event types)",
 "unit_of_analysis": "Instance: one named industrial enterprise in one article. An article naming two enterprises yields two rows; an article naming none yields one row with relevance = 0.",
 "research_question": "Which industrial sectors receive state-media attention, what is reported about their enterprises, and how has the pattern of reported leadership attention changed since 2012?",
 "researcher_interventions": [
  "Sector (v1: 8 codes) is recoded to four industrial sectors. Energy and mining are kept apart because the research question concerns sectoral attention and the two have different reporting patterns (power stations are construction and modernisation stories; mines are production stories).",
  "Non-industrial enterprises (farms, fisheries, transport, construction firms, utilities, services) are out of scope and coded relevance = 0 with a note, rather than kept in an 'other' residual.",
  "Event type (v1: 7 codes) is recoded to four. Technical innovation is folded into production; mobilisation campaigns, commendations and the residual are folded into political_ceremonial.",
  "leader_visit (kim_jong_un / premier_cabinet / none) and province are kept as proposed in v1."
 ],
 "variables": [
  {
   "name": "relevance",
   "label": "Relevance",
   "type": "binary",
   "definition": "1 if the article's subject is a named industrial enterprise (heavy industry, light industry, energy or mining) in North Korea; 0 otherwise.",
   "codes": [
    {
     "code": 1,
     "label": "relevant",
     "rule": "A specific industrial enterprise is named (기업소, 공장, 련합기업소, 발전소, 탄광, 광산 or equivalent) and is the subject or a main subject of the article."
    },
    {
     "code": 0,
     "label": "not relevant",
     "rule": "No specific enterprise is named; the enterprise is non-industrial (farm, fishery, transport, construction firm, utility, service, commerce); the enterprise is foreign; or 기업소 appears in the title for another reason. Note the reason in notes."
    }
   ]
  },
  {
   "name": "enterprise_name",
   "label": "Enterprise name",
   "type": "text",
   "definition": "The enterprise's name exactly as written in the article (Korean), without honorific prefixes.",
   "codes": []
  },
  {
   "name": "sector",
   "label": "Industrial sector",
   "type": "nominal",
   "definition": "The branch of industry to which the enterprise belongs, decided from its name first and from the article's description of its output second.",
   "codes": [
    {
     "code": "heavy_industry",
     "label": "Heavy industry",
     "rule": "Metal (steel, iron, non-ferrous), machine-building, locomotive and vehicle, shipbuilding, chemical and fertiliser, cement and building materials, electrodes and insulators, munitions and defence industry."
    },
    {
     "code": "light_industry",
     "label": "Light industry",
     "rule": "Consumer goods: textiles and clothing, footwear, food processing, daily necessities, cosmetics, paper, furniture, local-industry factories (지방공업공장) producing consumer goods."
    },
    {
     "code": "energy",
     "label": "Energy",
     "rule": "Power generation and supply: thermal and hydro power stations (발전소), grid and transmission enterprises. Coal mines are mining, not energy."
    },
    {
     "code": "mining",
     "label": "Mining",
     "rule": "Extraction of coal, ores, minerals and salt: 탄광, 광산, 련합탄광, 청년탄광, salt works (염전)."
    }
   ]
  },
  {
   "name": "event_type",
   "label": "Event type",
   "type": "nominal",
   "definition": "The main thing the article reports about the enterprise. One code per instance; apply the precedence rule in Decision_Rules when two events receive comparable attention.",
   "codes": [
    {
     "code": "leadership_visit",
     "label": "Leadership visit",
     "rule": "A senior official (Kim Jong Un, the Premier, a Cabinet or Party official) is reported as physically present at the enterprise: 현지지도, 현지료해, 시찰, 방문."
    },
    {
     "code": "production",
     "label": "Production and innovation",
     "rule": "Output results, plan fulfilment, productivity gains, new products, technical innovations, process improvements, scientific and technical achievements at the enterprise."
    },
    {
     "code": "construction_modernization",
     "label": "Construction and modernisation",
     "rule": "Building, expanding, renovating or re-equipping the enterprise or a production line; completion ceremonies for new facilities (준공식, 조업식)."
    },
    {
     "code": "political_ceremonial",
     "label": "Political and ceremonial",
     "rule": "Mobilisation campaigns, rallies, oaths and pledges; awards, commendations and titles; anniversaries and commemorations; and any other report that is not a visit, production or construction story."
    }
   ]
  },
  {
   "name": "leader_visit",
   "label": "Visiting official",
   "type": "nominal",
   "definition": "Who, if anyone, is reported as present at the enterprise in this article.",
   "codes": [
    {
     "code": "kim_jong_un",
     "label": "Kim Jong Un",
     "rule": "Kim Jong Un is reported as physically present (현지지도 etc.)."
    },
    {
     "code": "premier_cabinet",
     "label": "Premier or Cabinet/Party official",
     "rule": "The Premier, a Vice-Premier, a minister, or a Party secretary is reported as physically present; Kim Jong Un is not."
    },
    {
     "code": "none",
     "label": "No visit reported",
     "rule": "No senior official is reported as present. Quotations of the leader's instructions, letters or past visits do not count."
    }
   ]
  },
  {
   "name": "province",
   "label": "Province",
   "type": "nominal",
   "definition": "The province (도) or special city where the enterprise is located, from the text or from the enterprise's name.",
   "codes": [
    {
     "code": "pyongyang",
     "label": "Pyongyang",
     "rule": "평양시."
    },
    {
     "code": "pyongan_s",
     "label": "South Pyongan",
     "rule": "평안남도, incl. 남포 when stated as such; 순천, 안주, 덕천, 북창, 개천."
    },
    {
     "code": "pyongan_n",
     "label": "North Pyongan",
     "rule": "평안북도: 신의주, 구성, 룡천, 정주."
    },
    {
     "code": "chagang",
     "label": "Chagang",
     "rule": "자강도: 강계, 희천, 만포."
    },
    {
     "code": "ryanggang",
     "label": "Ryanggang",
     "rule": "량강도: 혜산, 삼지연."
    },
    {
     "code": "hamgyong_s",
     "label": "South Hamgyong",
     "rule": "함경남도: 함흥, 흥남, 단천, 신포."
    },
    {
     "code": "hamgyong_n",
     "label": "North Hamgyong",
     "rule": "함경북도: 청진, 김책, 회령, 무산."
    },
    {
     "code": "kangwon",
     "label": "Kangwon",
     "rule": "강원도: 원산, 문천."
    },
    {
     "code": "hwanghae_s",
     "label": "South Hwanghae",
     "rule": "황해남도: 해주."
    },
    {
     "code": "hwanghae_n",
     "label": "North Hwanghae",
     "rule": "황해북도: 사리원, 송림, 개성."
    },
    {
     "code": "nampo_rason",
     "label": "Nampo or Rason",
     "rule": "남포시 or 라선시 when named as a special city."
    },
    {
     "code": "unknown",
     "label": "Unknown",
     "rule": "Location cannot be determined from text or name; lower coder_confidence."
    }
   ]
  },
  {
   "name": "evidence_text",
   "label": "Evidence",
   "type": "text",
   "definition": "A verbatim excerpt (one or two sentences) from the article that justifies sector and event_type.",
   "codes": []
  },
  {
   "name": "coder_confidence",
   "label": "Coder confidence",
   "type": "ordinal",
   "definition": "The coder's confidence in the row's codes as a whole.",
   "codes": [
    {
     "code": 3,
     "label": "high",
     "rule": "Every code is stated or unambiguous in the text."
    },
    {
     "code": 2,
     "label": "medium",
     "rule": "One code is inferred (typically sector from a name, or province from a place name)."
    },
    {
     "code": 1,
     "label": "low",
     "rule": "Two or more codes inferred, or a boundary case under any rule."
    }
   ]
  }
 ],
 "decision_rules": [
  "Scope. An enterprise is in scope only if it belongs to one of the four industrial sectors. Farms, fisheries, transport and construction enterprises, utilities, shops and service units are out of scope: code relevance = 0 and write the reason in notes. This is a researcher decision for this study, not a property of the corpus.",
  "One row per named enterprise. Enterprises mentioned only as audience, contributors or comparison points, without their own reported event, receive no row.",
  "Sector by name first. Decide sector from the enterprise's name (제강소, 기계공장, 비료련합기업소 → heavy_industry; 방직공장, 식료공장, 신발공장 → light_industry; 발전소 → energy; 탄광, 광산 → mining) and from the article's description only when the name is uninformative.",
  "Event precedence. When two events receive comparable attention, code in this order: leadership_visit > construction_modernization > production > political_ceremonial. A visit during which production results are reported is a leadership_visit.",
  "Instructions are not visits. Quotations of the leader's words, letters, or references to a past visit do not make the current article a leadership_visit; code leader_visit = none unless presence in this article is reported.",
  "Officials at ceremonies. An official who attends a completion ceremony or rally at the enterprise is present: code leader_visit accordingly, and code event_type by precedence (construction_modernization for a completion ceremony unless the article is about the official's inspection).",
  "Unnamed but identifiable enterprises (e.g. 중요국방공업기업소) are relevance = 0; a dataset of named enterprises cannot carry them.",
  "Province from text, then from name, then unknown. Do not infer a province from the sector or from general knowledge about where an industry is concentrated.",
  "Evidence must be verbatim. evidence_text is copied from article_cleaned_body without alteration."
 ]
}