[{"data":1,"prerenderedAt":184},["ShallowReactive",2],{"topic-page:\u002Fhigh-school\u002Fmathematics-b\u002Fstatistical-inference\u002Fnull-alternative-and-significance":3},{"topic":4,"category":150,"seo":154,"breadcrumbs":159,"prerequisites":166,"nextTopics":176,"relatedTopics":183},{"id":5,"slug":6,"title":7,"summary":8,"canonicalPath":9,"categoryPath":10,"difficulty":11,"targetAudience":12,"estimatedMinutes":13,"generatedAt":14,"publishedAt":14,"updatedAt":14,"reviewStatus":15,"tags":16,"expectedKnowledge":21,"learningGoal":24,"blocks":25},"high-school-mathematics-b-statistical-inference-null-alternative-and-significance","null-alternative-and-significance","帰無仮説・対立仮説と有意性","帰無仮説と対立仮説を事前に定め、p値を有意水準と比べて棄却するか判断する検定の論理を学びます。","\u002Fhigh-school\u002Fmathematics-b\u002Fstatistical-inference\u002Fnull-alternative-and-significance","\u002Fhigh-school\u002Fmathematics-b\u002Fstatistical-inference","beginner","student_high",6,"2026-07-27","ai_generated",[17,18,19,20],"数学B","統計的な推測","仮説検定","有意水準",[22,23],"正規分布の面積","標本平均の標本分布","問いに対応する帰無仮説・対立仮説を設定し、p値と有意水準から適切な結論表現を選べる。",[26,31,41,68,74,84,105,111,117,131,140],{"id":27,"type":28,"title":7,"subtitle":29,"shortDefinition":30},"hero","HeroBlock","「差がない」と仮定して、結果の珍しさを測る","仮説検定では、まず差がないという帰無仮説を置きます。その下で観測結果以上の食い違いがどれほど珍しいかをp値で測り、事前の有意水準と比べます。",{"id":32,"type":33,"term":34,"definition":35,"plainExplanation":36,"keywords":37},"definition","DefinitionBlock","帰無仮説・対立仮説・p値","帰無仮説H₀は検討の基準、対立仮説H₁はデータから支持したい方向、p値はH₀の下での結果の珍しさです。","たとえば平均が500と異なるかを調べるならH₀: μ=500、H₁: μ≠500と置きます。H₀の下ではめったに出ないほど離れた標本平均ならH₀を棄却します。基準の有意水準αはデータを見る前に定めます。",[38,39,40,20],"H₀","H₁","p値",{"id":42,"type":43,"title":44,"lead":45,"formulas":46,"strategy":67},"formula","FormulaBlock","検定の判断規則","p値と事前に定めた有意水準αを比べます。",[47,58],{"label":48,"expression":49,"meaning":50,"symbols":51,"tip":57},"棄却の規則","p値≤α → H_0を棄却","H₀の下では観測結果が十分珍しいと判断し、H₁を支持する証拠があると結論します。",[52,54],{"symbol":40,"meaning":53},"H₀の下で観測結果以上に極端な結果が出る確率",{"symbol":55,"meaning":56},"α","事前に決める有意水準","等号の場合を含む規則か、授業や問題の定義に合わせて確認します。",{"label":59,"expression":60,"meaning":61,"symbols":62,"tip":66},"棄却できない規則","p値>α → H_0を棄却できない","H₀と食い違う証拠が設定した基準では十分でない、という判断です。",[63],{"symbol":64,"meaning":65},"棄却できない","H₀が正しいと証明した意味ではない","「H₀を採択して真と確定」とは書きません。","仮説、検定方向、有意水準を先に書き、最後は文脈の言葉で結論を戻します。",{"id":69,"type":70,"src":71,"alt":72,"caption":73},"diagram","DiagramBlock","\u002Fassets\u002Fhigh-school\u002Fmathematics-b\u002Fstatistical-inference\u002Fnull-alternative-and-significance\u002Fnull-alternative-and-significance.svg","帰無仮説の分布の両端に棄却域を示し、p値0.032と有意水準0.05を比較する図","両側検定では帰無仮説から両方向への食い違いを数えます。p=0.032はα=0.05以下なので、この規則ではH₀を棄却します。",{"id":75,"type":76,"steps":77,"title":83},"steps","StepBlock",[78,79,80,81,82],"母集団についての問いをH₀とH₁へ翻訳する","片側か両側か、有意水準αはいくつか事前に決める","H₀の下で検定統計量とp値を求める","p値とαを比べてH₀を棄却するか決める","効果の向き・大きさ・前提とともに文脈へ戻して述べる","仮説検定の5手順",{"id":85,"type":86,"columns":87,"rows":91,"summary":104},"wording","CompareBlock",[88,89,90],"結果","適切な表現","避ける表現",[92,96,100],[93,94,95],"p≤α","H₀を棄却する","H₁が100%真",[97,98,99],"p>α","H₀を棄却できない","H₀が証明された",[101,102,103],"小さいp","H₀との食い違いが大きい","H₀が真である確率が小さい","検定は証明ではなく、仮定したH₀と観測データの整合性を基準に沿って判断する手続きです。",{"id":106,"type":107,"scenario":108,"explanation":109,"result":110},"example","ExampleBlock","H₀の下で計算した両側p値が0.032、有意水準が5%である。","小数へそろえるとα=0.05です。0.032\u003C0.05なのでH₀を棄却します。この判断は「H₀の下なら観測結果以上の食い違いは約3.2%しか起こらない」という意味です。対立仮説を支持する統計的な証拠があると述べます。","5%有意水準でH₀を棄却します。",{"id":112,"type":113,"warningTitle":114,"message":115,"severity":112,"action":116},"warning","WarningBlock","p値は仮説の確率ではない","p=0.032は「H₀が正しい確率3.2%」ではありません。H₀が正しいと仮定した上で、今回以上に極端なデータが出る確率です。また有意でも、差が実用上大きいとは限りません。","p値の条件を「H₀の下で」と明記し、差の大きさや信頼区間も併せて読みます。",{"id":118,"type":119,"title":120,"questions":121},"quiz","QuizBlock","確認テスト",[122],{"question":123,"choices":124,"answerIndex":126,"choiceExplanations":127},"有意水準0.05でp値0.12を得たとき、適切な結論はどれですか。",[98,125,94],"H₀が正しいと証明された",0,[128,129,130],"正解です。0.12>0.05なので、この基準ではH₀と食い違う証拠が十分ではありません。","棄却できないことはH₀の真実性を証明することではありません。","棄却するのはp値が有意水準以下の場合です。",{"id":132,"type":133,"title":134,"points":135},"summary","SummaryBlock","まとめ",[136,137,138,139],"H₀は検定の基準、H₁は支持したい方向","p値はH₀の下で観測結果以上が出る確率","p値を事前の有意水準と比べる","棄却できないことはH₀の証明ではない",{"id":141,"type":142,"title":143,"nextTopics":144,"relatedKeywords":146},"next","NextActionBlock","次は母平均の検定を計算する",[145],"high-school-mathematics-b-statistical-inference-normal-test-for-population-mean",[147,148,149],"検定統計量","両側検定","母平均",{"path":10,"title":18,"description":151,"parentPath":152,"accentToken":153},"標本調査、標本分布、正規分布を用いた区間推定、仮説検定、信頼区間、有意水準を整理していくカテゴリです。","\u002Fhigh-school\u002Fmathematics-b","signal-blue",{"title":155,"description":156,"canonicalUrl":157,"ogImage":158},"帰無仮説・対立仮説と有意性 | 数学B 統計的な推測","仮説検定のH₀、H₁、p値、有意水準と、棄却・棄却できないの正しい読み方を学びます。","https:\u002F\u002Fzeqnilo.com\u002Fhigh-school\u002Fmathematics-b\u002Fstatistical-inference\u002Fnull-alternative-and-significance","https:\u002F\u002Fzeqnilo.com\u002Fassets\u002Fsite\u002Fdefault-og.svg",[160,163,164,165],{"title":161,"path":162},"高校","\u002Fhigh-school",{"title":17,"path":152},{"title":18,"path":10},{"title":7,"path":9},[167],{"id":168,"title":169,"summary":170,"canonicalPath":171,"categoryPath":10,"categoryTitle":18,"difficulty":11,"targetAudience":12,"estimatedMinutes":172,"publishedAt":14,"updatedAt":14,"reviewStatus":15,"tags":173},"high-school-mathematics-b-statistical-inference-confidence-level-width-and-sample-size","信頼度・区間幅・標本サイズ","信頼区間の誤差幅を分解し、信頼度、母標準偏差、標本サイズが区間の広さへどう影響するかを学びます。","\u002Fhigh-school\u002Fmathematics-b\u002Fstatistical-inference\u002Fconfidence-level-width-and-sample-size",5,[17,18,174,175],"信頼度","標本サイズ",[177],{"id":145,"title":178,"summary":179,"canonicalPath":180,"categoryPath":10,"categoryTitle":18,"difficulty":11,"targetAudience":12,"estimatedMinutes":13,"publishedAt":14,"updatedAt":14,"reviewStatus":15,"tags":181},"母平均の正規検定","帰無仮説の母平均から標本平均が標準誤差何個分離れたかをz値で測り、両側p値から検定します。","\u002Fhigh-school\u002Fmathematics-b\u002Fstatistical-inference\u002Fnormal-test-for-population-mean",[17,18,149,182],"正規検定",[],1785565321845]