モジュール26：図表・複数資料の読解

入試での出題分析

出題形式と難易度

項目	評価
難易度	★★★★★ 発展
分量	多い（図表3-5個 + 本文800-1200語）
処理速度	厳しい（15-20分/大問）
統合度	極めて高い（言語+視覚+論理）

頻出パターン

早慶・旧帝大

• 経済・社会科学分野の統計データ（グラフ、表）を含む長文が出題され、データの正確な読解と本文の議論との統合的理解が問われる。（例：早稲田大学法学部・政治経済学部、慶應義塾大学経済学部）
• 複数の図表（例：傾向を示すグラフと、その内訳を示す表）を比較・対照させ、両者から導き出される結論や、両者の間に存在する矛盾について説明を求める設問。
• 本文の記述と図表のデータが一致するかどうかを問う内容一致問題。
• 図表から読み取れる傾向や事実に基づき、その背後にある原因や、将来の結果を推論させる記述問題。

難関国公立

• 自然科学分野（特に生物学、地学、環境科学）の実験結果や観測データを示す図表を含む長文。
• 図表に示されたデータから、実験の仮説や結論の妥当性を評価させる批判的読解問題。
• 図表の視覚的表現（軸の操作など）の問題点を指摘させたり、異なる表現方法の利点・欠点を論じさせたりする、高度なデータリテラシーを問う設問。

差がつくポイント

1. 統語処理の正確性: increase by/to の区別、respectively の正確な対応付け、the former/latter の照応関係など、数値を扱う表現の統語的構造を精密に処理できるか。
2. 意味解釈の深さ: 数値の単位・基準値・変化の大きさを文脈に沿って評価し、平均値の裏にある分散や異常値の意味を読み解き、データから妥当な推論を導出できるか。
3. 複数資料の統合能力: 複数の図表が示す情報の一致、矛盾、補完関係を正確に識別し、それらを統合して一つの論証として再構築できるか。
4. 批判的・語用論的視点: データの選択、グラフ形式の選択、軸の操作といった、筆者の修辞的意図やバイアスを見抜き、提示された情報の信頼性を客観的に評価できるか。
5. 処理の優先順位判断: 情報の階層性（主要情報 vs 詳細情報）を素早く見抜き、制限時間内で解答に必要な情報を効率的に取捨選択できるか。

演習問題

試験時間: 60分 / 満点: 100点

第１問（25点）

次の文章と図表を読み、設問に答えよ。

The debate over the effectiveness of carbon pricing mechanisms in reducing greenhouse gas emissions has intensified as countries implement various policy approaches. Figure 1 displays the relationship between carbon price levels ($/ton CO2) and emission reductions (% change from baseline) across 35 jurisdictions that have implemented carbon pricing between 2010 and 2023. The scatter plot reveals a positive correlation (r=0.72), with jurisdictions imposing prices above $50/ton achieving average emission reductions of 18-22%, while those with prices below $30/ton show reductions of only 5-8%. However, as Table 1 demonstrates, the relationship is complicated by numerous confounding variables.

Table 1: Carbon Pricing Outcomes by Jurisdiction Type

Jurisdiction Type	Average Price ($/ton)	Emission Reduction (%)	GDP Growth (% annual)	Policy Comprehensiveness Score (out of 10)
Nordic Countries	68	21	2.1	8.7
Other EU	45	14	1.8	7.2
North America	35	9	2.4	5.8
East Asia	28	7	3.2	4.9
Emerging Markets	18	4	4.1	3.2

Figure 2 tracks emission trajectories for three representative countries over 2015-2023: Sweden (which maintained a carbon price above $120/ton), Canada (which gradually increased its price from $15 to $50/ton), and Japan (which implemented a price of $3/ton with limited coverage). Sweden’s emissions declined by 27%, Canada’s by 12%, and Japan’s by 3% over this period. Yet, as Chart 1 illustrates, when emissions are calculated on a consumption basis—accounting for goods imported from abroad—Sweden’s reduction shrinks to 14%, Canada’s to 8%, and Japan’s holds at 3%, revealing how international trade allows countries to externalize emissions.

(1) According to Table 1, what is the relationship between carbon price levels and GDP growth rates across jurisdiction types, and what might explain this relationship? (8 points)

(2) The text states that Figure 1 shows a correlation of r=0.72 between carbon prices and emission reductions. Based on the information provided in the passage, explain two reasons why this correlation might not indicate a simple causal relationship where price directly causes reduction. (9 points)

(3) Compare the production-based and consumption-based emission reductions for Sweden as presented in the text. What does the difference between these two measures (27% vs 14%) suggest about the nature of Sweden’s emission reductions, and what are the implications for evaluating the true effectiveness of its carbon pricing policy? (8 points)

第２問（25点）

次の文章と図表を読み、設問に答えよ。

Educational mobility—the extent to which children’s educational attainment exceeds that of their parents—has emerged as a critical indicator of social dynamism and opportunity. Figure 3 presents data on intergenerational educational mobility across 28 OECD countries, measuring the percentage of individuals aged 25-44 who have attained higher education levels than their parents. The data reveal substantial variation, ranging from 65% in South Korea to 28% in Germany, with an OECD average of 47%.

Table 2 disaggregates mobility rates by parental education level, revealing a troubling pattern. Among children of parents without high school completion, only 15% attain university degrees, compared to 68% of children whose parents hold graduate degrees—a 53-percentage-point gap that has widened from 41 percentage points in 1980. Chart 2 correlates these mobility rates with public education expenditure as a percentage of GDP, showing a modest positive relationship (r=0.41), though several outliers complicate the picture: Finland achieves 58% mobility with 6.2% of GDP spent on education, while the United States achieves only 43% mobility despite spending 6.1% of GDP.

Figure 4 tracks mobility trends over time for five countries (1980-2020 in decade intervals), revealing that while absolute educational attainment has risen universally—the percentage holding tertiary degrees increased from 15% to 45% on average—relative mobility has stagnated or declined in four of the five countries examined. This paradox suggests that as higher education has expanded, its distributional benefits have become increasingly unequal.

(1) Based on Table 2, calculate the ratio of university attainment rates between children of graduate-degree holders and children of parents without high school completion. What does this ratio, and its change since 1980, indicate about educational inequality? (7 points)

(2) The text notes that Finland and the United States spend similar percentages of GDP on education (6.2% vs 6.1%) but achieve very different mobility outcomes (58% vs 43%). Drawing on information from the passage and your general understanding, propose two distinct explanations for this disparity. (10 points)

(3) Explain the paradox described in the final paragraph: how is it possible for absolute educational attainment to rise for everyone while relative mobility stagnates or declines? What does this paradox suggest about the relationship between educational expansion and social equality? (8 points)

第３問（25点）

次の文章と図表を読み、設問に答えよ。

The relationship between automation and employment has generated intense debate. Figure 5 displays employment trends across three skill categories—high-skill (professional/managerial), middle-skill (clerical/manufacturing), and low-skill (service/manual labor)—from 1980 to 2023, revealing a pronounced pattern of labor market polarization. High-skill employment expanded from 28% to 41% of total employment, low-skill employment grew modestly from 19% to 23%, while middle-skill employment contracted dramatically from 53% to 36%, a phenomenon economists term “hollowing out.”

Table 3 presents data on automation exposure and employment changes by occupation. Occupations with high routine task intensity (tasks codifiable into algorithms) experienced employment declines of 32% on average, while occupations with low routine intensity saw gains of 27%. However, Chart 3 reveals that the relationship between automation and wage changes is more complex: high-skill occupations facing automation pressure (e.g., financial analysts) experienced wage increases of 35%, whereas low-skill occupations with similar exposure (e.g., checkout clerks) saw wages decline by 8%.

Figure 6 projects future automation impact under three scenarios through 2040: a low-disruption scenario predicts 18% of jobs will be transformed, a medium-disruption scenario predicts 39%, and a high-disruption scenario predicts 61%. The distribution of impact across skill levels varies dramatically, with the high-disruption scenario threatening not only middle-skill but also certain high-skill occupations previously considered automation-resistant.

(1) According to Figure 5, calculate the absolute change in percentage points for each of the three skill categories between 1980 and 2023. Which skill category experienced the largest absolute decline, and what does this “hollowing out” pattern suggest about the nature of recent technological change? (8 points)

(2) Chart 3 shows that high-skill and low-skill occupations experience opposite wage effects from automation. Drawing on economic principles, explain why automation would increase wages in some high-skill occupations while decreasing them in many low-skill occupations. (Hint: consider the concepts of complement and substitute). (9 points)

(3) Compare the future high-disruption scenario in Figure 6 with the historical pattern shown in Figure 5. What would be the most significant difference in the pattern of job displacement, and why would this difference make future policy responses more challenging than in the past? (8 points)

第４問（25点）

次の文章と図表を読み、設問に答えよ。

The global distribution of COVID-19 vaccine access during 2021-2022 exposed profound inequities. Figure 7 displays cumulative vaccination rates (% of population fully vaccinated) as of December 2022 across income groups: high-income countries averaged 76%, upper-middle-income 68%, lower-middle-income 52%, and low-income countries a mere 21%. This 55-percentage-point gap persisted despite international initiatives like COVAX.

Table 4 examines the relationship between vaccination rates and mortality. Countries achieving >70% vaccination rates by mid-2021 experienced average mortality of 120 deaths per 100,000 population over the subsequent 18 months, while countries below 40% vaccination experienced 340 deaths per 100,000—a 2.8-fold difference. However, as Chart 4 demonstrates, this cannot be attributed solely to vaccination: countries with robust healthcare systems (measured by hospital beds and physicians per 1,000 population) achieved lower mortality even at similar vaccination rates.

Figure 8 tracks vaccine production capacity and distribution monthly from January 2021 to December 2022. It shows that while global production capacity expanded from 300 million to 1.8 billion doses monthly by late 2021, distribution remained heavily concentrated in high-income countries through mid-2022. Not until Q3 2022 did low-income countries begin receiving substantial supplies, by which point high-income countries were administering booster doses while many healthcare workers in poor countries remained unvaccinated—a sequence the WHO termed “vaccine apartheid.”

(1) Using data from Table 4, calculate the mortality rate difference (in deaths per 100,000) between high- and low-vaccination countries. Given the information in Chart 4, explain why it would be inaccurate to attribute this entire difference to vaccination alone. (8 points)

(2) Figure 7 shows a 55-point vaccination gap, while Figure 8 shows that global production capacity reached 1.8 billion doses/month by late 2021. Given these two facts, was the persistent vaccine gap a problem of insufficient production or something else? Explain what Figure 8 suggests were the true barriers to equitable distribution. (9 points)

(3) The text describes a situation where high-income countries administered booster doses while healthcare workers in poor countries remained unvaccinated. From both an ethical and a global public health perspective, evaluate this distribution pattern. (8 points)

解答・解説

難易度構成

難易度	配点	大問
標準 (Standard)	31点	第１問(1), 第２問(1), 第３問(1), 第４問(1)
発展 (Advanced)	36点	第１問(2), 第２問(2), 第３問(2), 第４問(2)
難関 (Applied)	33点	第１問(3), 第２問(3), 第３問(3), 第４問(3)

結果の活用

得点	判定	推奨アクション
80点以上	A	過去問演習へ進む。特に早慶の複合資料問題を集中的に解き、時間配分戦略を確立する。
60-79点	B	弱点分野を補強後、再挑戦。特に「意味層」で扱った数値からの推論導出や、「語用層」での筆者の意図の読解を強化する。
40-59点	C	講義編を復習後、再挑戦。「統語層」の数値表現の処理と、「意味層」のグラフの種類別特性の理解を固める。
40点未満	D	講義編を再学習。特に「統語層」の図表説明文の構造分析から始め、基礎を徹底的に再構築する。

第１問 解答・解説

【戦略的情報】

項目	内容
出題意図	表とグラフから相関関係と交絡要因を読み解き、異なる測定基準の意味を考察する能力を問う。
難易度	発展
目標解答時間	15分

【思考プロセス】

状況設定

残り時間45分。4大問中の最初として、データリテラシーの基礎が問われている。表と本文の両方から情報を抽出し、統合する必要がある。受験生は相関関係を因果関係と誤解しやすいが、本文は交絡変数の存在を明示している。

レベル1：初動判断

(1) Table 1の「Average Price」列と「GDP Growth」列の関係を見る。数値を縦に比較し、パターンを識別する。
(2) 設問は「相関が因果を意味しない理由」を問うている。本文中のconfounding variables、Table 1のPolicy Comprehensiveness Scoreに注目する。
(3) 本文中のスウェーデンの数値を正確に抽出する：production-based (27%)とconsumption-based (14%)。

レベル2：情報の取捨選択

(1) 価格と成長率の間に「負の相関」があることを示す数値ペアを確認。Nordic（価格68、成長2.1）とEmerging Markets（価格18、成長4.1）を対比させる。
(2) Figure 1の相関だけでなく、Table 1全体の構造を見る。政策包括性スコアが価格と連動している点に着目。
(3) 27%と14%の差である13ポイントが何を意味するかを考察する。本文のexternalize emissionsという表現がヒント。

レベル3：解答構築

(1) 負の相関を指摘した上で、経済発展段階という「第三の変数」の可能性を考察する。
(2) 二つの理由として、①政策の包括性という交絡変数の存在、②削減意欲の高い国が高価格を導入する「自己選択バイアス」を挙げる。
(3) 「炭素リーケージ（排出の外部化）」の概念を用いて、見かけ上の削減と実質的な削減の乖離を説明する。

判断手順ログ

手順1：Table 1の列を縦に比較し、価格↑と成長率↓の負の相関を確認。
手順2：この負の相関の因果的説明として、先進国/新興国という経済発展段階を第三変数として導入。
手順3：Figure 1の相関（r=0.72）に対し、Table 1の「Policy Comprehensiveness」が交絡変数として機能する可能性を指摘。
手順4：自己選択バイアス（意欲の高い国→高価格導入）を第二の理由として追加。
手順5：スウェーデンの生産ベース27%と消費ベース14%の差を「炭素リーケージ」と結びつけ、政策評価への含意を論述。

【解答】

(1) Table 1 shows an inverse relationship between average carbon price and GDP growth rate: jurisdictions with higher prices (e.g., Nordic Countries at $68) exhibit lower growth (2.1%), while those with lower prices (e.g., Emerging Markets at $18) show higher growth (4.1%). This correlation likely does not mean high prices cause low growth; rather, it probably reflects that high-income, mature economies (which can afford high carbon prices) naturally have lower GDP growth rates, whereas emerging markets are in a phase of rapid, energy-intensive growth.

(2) First, the correlation may be confounded by policy comprehensiveness. Table 1 shows that jurisdictions with high prices, like Nordic countries, also have high “Policy Comprehensiveness” scores (8.7/10). The emission reduction may be a result of this comprehensive package of policies (e.g., subsidies for renewables, efficiency standards) rather than the price mechanism alone. Second, a self-selection bias may be at play. Jurisdictions that are politically and economically capable of, and committed to, ambitious climate action are more likely to implement high carbon prices in the first place. Therefore, the correlation may reflect this underlying commitment rather than a direct causal effect of the price itself.

(3) Sweden’s production-based emission reduction (27%) is nearly double its consumption-based reduction (14%). The 13-percentage-point difference suggests that a significant portion of Sweden’s apparent success is due to “carbon leakage”—offshoring carbon-intensive production to other countries while importing the finished goods. This implies that its domestic carbon price is partially effective at reducing domestic production emissions but may be less effective at reducing Sweden’s global carbon footprint. For a true evaluation of policy effectiveness, consumption-based accounting is crucial as it prevents countries from claiming climate progress by simply exporting their emissions.

【解答のポイント】

正解の論拠: (1)負の相関の指摘と、交絡要因（経済発展段階）の考察。(2)政策包括性スコアと自己選択バイアスという、二つの具体的な交絡要因を本文・表から読み取って指摘。(3)生産ベースと消費ベースの差を「炭素リーケージ」と結びつけ、政策評価への含意を明確に論じている。

誤答の論拠: (1)単に「逆相関がある」と述べるだけで、その理由を考察しない。(2)「相関は因果ではない」という一般論に終始し、本文中の具体的な根拠を示せない。(3)13ポイントの差の計算はできても、それが何を意味し、なぜ重要なのかを説明できない。

【再現性チェック】

この解法が有効な条件: 散布図で相関関係が示され、表で複数の変数が並列されている複合資料問題。「相関≠因果」を問う設問では、本文・表中の交絡変数候補を探す。

【参照】

• [M26-意味] └ 数値データからの推論の導出
• [M26-語用] └ データの選択的提示と省略の意図を読む

第２問 解答・解説

【戦略的情報】

項目	内容
出題意図	教育格差のデータを多角的に分析し、「絶対的改善」と「相対的格差拡大」というパラドックスの構造を理解する能力を問う。
難易度	難関
目標解答時間	15分

【思考プロセス】

状況設定

残り時間30分。教育社会学的なテーマで、数値の比較だけでなく、その社会的意味の解釈が求められる。特に(3)の「パラドックス」は、単なるデータ読解を超えた概念的理解が必要。

レベル1：初動判断

(1) Table 2から、親の学歴が「大学院卒」と「高校未卒」の子供の大学進学率（68%と15%）を抜き出す。比率計算と1980年との比較。
(2) Chart 2の「外れ値」としてフィンランドと米国のデータに着目。支出はほぼ同じ、成果は異なる。
(3) 最終段落の「パラドックス」の定義を確認：絶対的達成度は上昇、相対的移動性は停滞・低下。

レベル2：情報の取捨選択

(1) 68% ÷ 15% = 4.53倍。1980年の格差（41ポイント）から2020年の格差（53ポイント）への拡大を確認。
(2) 「支出の量」が同じでも「支出の質・配分」が異なる可能性。教育以外の社会的要因（所得格差、福祉制度）の可能性。
(3) 高等教育の「拡大」と「階層化」の同時進行。全員が進学しやすくなっても、有利な家庭の子供はより上位に進む。

レベル3：解答構築

(1) 比率の意味を「機会の不平等」と結びつける。格差拡大を世代間不平等の深刻化として解釈。
(2) フィンランドの平等主義的配分と米国の地域格差という対比。教育外の社会構造（福祉国家 vs 市場主義）という対比。
(3) 「相対的移動性」の停滞は、教育が階層を固定化するメカニズムの存在を示唆。

判断手順ログ

手順1：Table 2から68%と15%を抽出し、比率4.53を計算。
手順2：1980年の41ポイント差と現在の53ポイント差を比較し、格差「拡大」を確認。
手順3：フィンランドと米国の比較から、「支出の質・配分」仮説を構築。
手順4：教育外要因として、北欧型福祉国家と米国型市場経済の対比を導入。
手順5：「絶対」と「相対」の概念を区別し、教育拡大が階層化を伴うメカニズムを説明。

【解答】

(1) The ratio is 68% / 15% ≈ 4.53. This means that children of parents with graduate degrees are over 4.5 times more likely to attain a university degree than children of parents who did not complete high school. This stark ratio indicates severe intergenerational inequality in educational opportunity. The fact that the gap has widened from 41 to 53 percentage points since 1980 suggests that this educational stratification is intensifying over time, meaning family background has become an even more powerful determinant of a child’s educational success.

(2) Two possible explanations: First, the disparity may stem from the quality and distribution of spending, not the total amount. Finland might allocate its resources more equitably across all schools and provide robust, universal support systems (e.g., high-quality early childhood education, school welfare services) that mitigate socioeconomic disadvantage. In contrast, U.S. education funding often relies on local property taxes, leading to vast spending disparities between wealthy and poor districts, which exacerbates inequality. Second, factors outside the formal education system can play a decisive role. Finland’s lower overall income inequality and comprehensive social safety net may create a society where parental income is a less critical factor for a child’s success. Conversely, high income inequality and a weaker safety net in the U.S. may mean that affluent families can invest far more in supplemental resources (tutoring, extracurriculars, enrichment) that perpetuate their advantage.

(3) The paradox arises because as higher education expands, it also becomes more stratified. Absolute attainment rises because a larger percentage of the entire population gains access to some form of tertiary education. However, relative mobility stagnates or declines because privileged families adapt their strategies to maintain their advantage. As a bachelor’s degree becomes more common, they ensure their children access more elite universities or more lucrative fields of study (e.g., medicine, law, engineering). Consequently, while a child from a disadvantaged background might now get a degree (an absolute improvement over their parents), the child from an advantaged background gets a betterdegree, preserving or even widening the relative gap in life outcomes. This suggests that simply expanding educational access, without addressing underlying inequalities and the hierarchical nature of the system, does not guarantee greater social equality.

【解答のポイント】

正解の論拠: (1)比率の計算と、それが示す「世代間格差」およびその「悪化」を明確に言語化。(2)「支出の質」と「教育以外の社会要因」という二つの異なる論点を立て、具体例を交えて説明。(3)高等教育の「拡大」と「階層化」という二つの側面を対比させ、パラドックスが生じるメカニズムを論理的に説明している。

誤答の論拠: (1)計算だけで解釈がない。(2)「教育の質が違うから」といった抽象的な説明に終始する。(3)「絶対」と「相対」の違いを説明できず、単に文章を言い換えるだけ。

【再現性チェック】

この解法が有効な条件: 社会科学分野で、複数の国や集団の比較データが示され、単純な相関では説明できない「外れ値」や「パラドックス」が存在する問題。

【参照】

• [M26-意味] └ 比較基準の特定と相対的位置づけ
• [M26-談話] └ 複合資料テクスト全体の論証構造の評価

第３問 解答・解説

【戦略的情報】

項目	内容
出題意図	労働市場の「二極化」の構造をデータから読み解き、自動化がスキルレベルによって異なる影響（補完 vs 代替）を与えるメカニズムを理解する能力を問う。
難易度	難関
目標解答時間	15分

【思考プロセス】

状況設定

残り時間15分。経済学的な概念（補完財・代替財）の理解が求められる。設問のヒントを活用することが重要。

レベル1：初動判断

(1) Figure 5から1980年と2023年の各技能レベルの雇用シェアを読み取り、差を計算する。
(2) Chart 3の、高技能と低技能で賃金変化の方向が逆転している点に着目。ヒント「complement and substitute」を使う。
(3) Figure 6（未来）とFigure 5（過去）のパターンを比較する。

レベル2：情報の取捨選択

(1) High-skill: 41-28=+13, Middle-skill: 36-53=-17, Low-skill: 23-19=+4。最大減少は中技能の-17。
(2) 高技能：自動化は「道具（補完財）」→生産性向上→賃金上昇。低技能：自動化は「競争相手（代替財）」→需要減少→賃金低下。
(3) 過去：中技能が破壊され、高技能・低技能に移動。未来：高技能も影響を受ける可能性。

レベル3：解答構築

(1) 「ホローイングアウト」を「定型業務（routine task）」の自動化と結びつける。
(2) 経済学の「補完」と「代替」の概念を正確に用いて説明する。
(3) 過去と未来の違いは「高技能職の安全性」。これが失われることで「スキルアップ」という政策が機能不全に陥る。

判断手順ログ

手順1：Figure 5の数値を抽出し、差を計算。最大減少が中技能であることを確認。
手順2：「ホローイングアウト」を定型業務の自動化というメカニズムで説明。
手順3：Chart 3の賃金格差を「補完」と「代替」の概念で解釈。
手順4：Figure 6の高混乱シナリオが高技能にも及ぶ点を過去のパターンと対比。
手順5：政策対応の困難さを「再教育戦略の限界」として論述。

【解答】

(1) The absolute changes in percentage points are: High-skill: 41% – 28% = +13 points; Middle-skill: 36% – 53% = -17 points; Low-skill: 23% – 19% = +4 points. The middle-skill category experienced the largest absolute decline (-17 points). This “hollowing out” pattern suggests that technological change has been skill-biased in a specific way: it has primarily automated and replaced routine tasks, both manual and cognitive, that were characteristic of middle-skill jobs like factory work and clerical positions. It did not replace non-routine tasks, complementing the abstract, creative tasks of high-skill workers and having little effect on the non-routine manual tasks of low-skill service workers.

(2) This disparity reflects whether automation acts as a complement to or a substitute for human labor. For high-skill workers (e.g., financial analysts, radiologists), AI and software are powerful tools that augment their abilities, making them more productive. Automation handles routine data processing, freeing them to focus on higher-level interpretation, strategy, and decision-making. This increases their marginal productivity and, consequently, the demand for their skills and their wages. For many low-skill workers performing routine tasks (e.g., checkout clerks, assembly workers), automation is a direct substitute. A self-checkout machine or an assembly robot can perform the same tasks, often more cheaply and efficiently. This reduces the demand for human labor in those roles, putting downward pressure on wages or eliminating the jobs entirely.

(3) The most significant difference is that the high-disruption scenario projects significant transformation and displacement even within the high-skill occupations, which were the primary beneficiaries of technological change in the historical pattern shown in Figure 5. In the past, the “hollowing out” of the middle meant displaced workers could theoretically “skill up” into the expanding high-skill sector. This makes future policy responses more challenging because the traditional solution of “upskilling” and “retraining” may no longer be a viable path to safety. If even highly educated professional jobs are subject to automation, it raises fundamental questions about where displaced workers can transition to, potentially requiring more radical policy solutions like universal basic income or large-scale public employment programs, rather than just educational reforms.

【解答のポイント】

正解の論拠: (1)正確な計算と、「ホローイングアウト」を「定型業務の自動化」というメカニズムと結びつけて説明。(2)「補完(complement)」と「代替(substitute)」という経済学の概念を正しく用い、論理的に説明。(3)過去と未来のパターンの本質的な違いを指摘し、政策対応の困難さを明確に論じている。

誤答の論拠: (1)計算ミス。または「ホローイングアウト」を単に「真ん中が減った」としか説明できない。(2)補完・代替の概念を使わずに、現象面の説明に終始する。(3)未来のほうが「大変だ」というだけで、具体的に説明できない。

【再現性チェック】

この解法が有効な条件: 経済・労働市場に関するデータで、スキルレベルや職種によって影響が異なるパターンが示されている問題。

【参照】

• [M26-意味] └ グラフの種類に応じた意味特性の理解
• [M26-談話] └ 複合資料テクストにおけるマクロ構造の把握

第４問 解答・解説

【戦略的情報】

項目	内容
出題意図	複数の図表を統合して、ある社会問題（ワクチン不平等）の多面的な原因と結果を分析し、倫理的評価を行う能力を問う。
難易度	難関
目標解答時間	15分

【思考プロセス】

状況設定

最終問題。複数の図表を統合し、因果関係の評価と倫理的判断を求められる。Chart 4の交絡変数、Figure 8の生産と配分の区別が重要。

レベル1：初動判断

(1) Table 4から死亡率（340と120）を読み取り、差を計算。Chart 4の「医療制度の質」に注目。
(2) Figure 8から「生産能力は十分だった」こと、「配分が高所得国に集中した」ことを読み取る。
(3) 「倫理的観点」と「公衆衛生的観点」の二つの視点で評価する。

レベル2：情報の取捨選択

(1) 340 – 120 = 220 の差。Chart 4は交絡変数（医療制度）の存在を示唆。
(2) 問題は「生産」ではなく「配分」。事前購入契約、知財、インフラの障壁。
(3) 倫理：功利主義・平等主義に反する。公衆衛生：変異株リスク、自己破壊的。

レベル3：解答構築

(1) 交絡変数の存在により、220の差全てをワクチンに帰属させることは不正確と論じる。
(2) 「生産 vs 配分」の問題を明確に切り分け、配分の政治的・経済的障壁を指摘。
(3) 二つの観点を区別し、それぞれの論点を明確に述べる。

判断手順ログ

手順1：Table 4から340と120を抽出し、差220を計算。
手順2：Chart 4を根拠に、医療制度という交絡変数の存在を指摘。
手順3：Figure 8から生産能力（1.8B doses/month）と配分の偏りを読み取る。
手順4：配分の障壁として、事前購入契約、知財、インフラを列挙。
手順5：倫理的評価（不公正）と公衆衛生的評価（変異株リスク）を区別して論述。

【解答】

(1) The mortality rate difference is 340 – 120 = 220 deaths per 100,000 population. It would be inaccurate to attribute this entire difference to vaccination alone because, as Chart 4 demonstrates, there is a confounding variable: the quality of healthcare systems. Chart 4 shows that even at similar vaccination rates, countries with more robust healthcare systems (more beds, more physicians) achieved lower mortality. This means that part of the 220-death difference is likely due to the fact that high-vaccination countries also tend to be wealthy countries with better hospitals and healthcare, which can save more lives irrespective of vaccination status. Vaccination is a major factor, but not the only one.

(2) The persistent vaccine gap was a problem of distribution and access, not insufficient production. Figure 8 clearly shows that global production capacity was vast (1.8 billion doses/month), theoretically enough to vaccinate the world population quickly. However, it also shows that this supply was “heavily concentrated in high-income countries.” This suggests the true barriers were political and economic, not technical. They likely included: 1) Advance Purchase Agreements, where wealthy nations bought up the majority of the supply before it was even produced, leaving little for others. 2) Intellectual Property restrictions that prevented widespread local manufacturing in lower-income countries. 3) Logistical and infrastructural challenges (e.g., cold chain requirements, last-mile delivery systems) in poorer nations.

(3) From an ethical perspective, the pattern is indefensible under most frameworks. A utilitarian approach, aiming to save the most lives globally, would have prioritized vaccinating all high-risk individuals worldwide (like healthcare workers) before giving non-essential boosters to low-risk individuals in wealthy nations. It violates principles of fairness and global equity. From a global public health perspective, the pattern was strategically flawed and self-defeating. Leaving vast populations unvaccinated creates a breeding ground for new, potentially more dangerous or vaccine-evasive variants of the virus. The emergence of such a variant would threaten everyone, including the vaccinated populations in high-income countries. This demonstrates that in a pandemic, national interest is inextricably linked to global health equity; “vaccine apartheid” is not only unethical but also a risk to all. The dilemma highlights the tension between national political obligations and the demands of global solidarity.

【解答のポイント】

正解の論拠: (1)差を正確に計算し、Chart 4を根拠に「交絡変数」の存在を指摘して、単純な因果帰属を否定。(2)「生産」の問題ではなく「配分」の問題であると明確に切り分け、その具体的な障壁を指摘。(3)「倫理」と「公衆衛生」の二つの観点を明確に区別し、それぞれの論点を論じている。

誤答の論拠: (1)交絡変数の存在に言及せず、220という差を全てワクチンの効果としてしまう。(2)問題の所在を「生産か配分か」で明確にできない。(3)倫理と公衆衛生の区別なく、一面的な感想に留まる。

【再現性チェック】

この解法が有効な条件: 社会問題に関する複数の図表が提示され、因果関係の評価と規範的判断（倫理的評価）の両方が求められる問題。

【参照】

• [M26-談話] └ 図表と本文の論理的連結の分析
• [M26-談話] └ 複合資料テクスト全体の論証構造の評価

体系的接続

• [M26-統語] └ 図表参照表現の統語的処理能力を、演習問題の正確な読解に応用する
• [M26-意味] └ 数値データの多角的解釈能力を、複数の図表が示す情報の統合に応用する
• [M26-語用] └ データの選択的提示や軸操作への批判的視点を、設問の意図を見抜く作業に応用する