References, data notes & research quality参考文献、数据说明与研究质量

What the source says: the pie chart on p.4 labels sectors "Positive:61 (61%)", "Neutral:27 (27%)", "Negative:12 (12%)".原文表述：第 4 页饼图标注为“Positive:61 (61%)”“Neutral:27 (27%)”“Negative:12 (12%)”。

Issue: with 127 total entries, 61 entries would equal 48.0%, not 61%. A value cannot simultaneously be a count and a percentage of the same base unless the base is 100. The three labels sum to 127 only if read as counts; they sum to 100 only if read as percentages.问题：在 127 条总量下，61 条应等于 48.0%，而非 61%。同一基数下，一个数值不可能同时是计数与百分比（除非基数恰为 100）。按计数解读三者合计 127，按百分比解读合计 100，二者只能取其一。

Treatment: treated as percentages throughout the site, matching the table's "Share" column. The count interpretation is not used anywhere. Flagged at the chart and in Table 3.2.处理：全站按百分比使用，与表格“占比”列保持一致；任何位置均未采用计数解读。已在图表与表 3.2 处标注。

What the source says: overall sentiment 61 / 27 / 12; media 72 / 21 / 7; owner community 48 / 22 / 30.原文表述：整体情绪 61 / 27 / 12；媒体 72 / 21 / 7；车主社群 48 / 22 / 30。

Issue: no single channel weight reproduces the overall figures. A weight of 54.2% media yields 61% positive, but then negative would be 17.5% rather than 12%. A weight that yields 12% negative (78.3% media) yields 66.8% positive instead of 61%. The overall mix therefore cannot be a weighted average of the two channel mixes as published.问题：不存在任何单一渠道权重可复现整体数值。媒体权重 54.2% 时正面为 61%，但负面会变为 17.5% 而非 12%；若要使负面为 12%（媒体权重 78.3%），正面则为 66.8% 而非 61%。因此整体构成不能是两组渠道比例的加权平均。

Treatment: all three sets retained as published. No weighted average is computed or implied. The gap column in Table 4.1 is a simple difference, not a reconciliation.处理：三组数据均按原文保留，不计算也不暗示任何加权平均。表 4.1 的差值列仅为简单相减，并非勾稽结果。

What the source says: Table 6.1 gives Germany 67% hybrid / 33% EV and Italy 60% EV / 40% hybrid. The chart on p.8 ("EV vs Hybrid Mentions Share by Market") shows Germany at roughly 33% EV with a near-zero hybrid bar, France at 100% EV, Italy at 60% EV with a very small hybrid bar.原文表述：表 6.1 给出德国混动 67% / 纯电 33%，意大利纯电 60% / 混动 40%。第 8 页图表（“各市场纯电与混动提及占比”）显示德国约 33% 纯电、混动柱接近零，法国 100% 纯电，意大利 60% 纯电、混动柱极低。

Issue: the hybrid shares in the table (67% and 40%) do not appear in the chart (approximately 2–5%). Either the table's hybrid shares are misstated, or the chart's series is mislabelled or mis-scaled. The report does not resolve which.问题：表中的混动占比（67% 与 40%）在图表中并未出现（约 2–5%）。可能原因包括表格混动占比有误，或图表系列标注/刻度有误；报告未给出结论。

Treatment: the table is treated as the primary record and plotted in Module 6; the p.8 chart is described but not reproduced. No hybrid value is averaged between the two.处理：以表格为主记录，并在模块 6 中绘制；第 8 页图表仅作说明，不予复现。两份数据之间不做任何平均。

What the source says: "18–22% real world highway range deficit against official WLTP figures" (negative drivers, [1]); "underperforms official laboratory specifications by 15–25%" (key finding 4).原文表述：“真实高速续航相对官方 WLTP 数值缺口 18–22%”（负面动因，[1]）；“低于官方实验室规格 15–25%”（核心发现 4）。

Issue: two ranges overlap but differ in both bounds. The wider range may aggregate charging as well as range, or may derive from a different test set; the report does not say.问题：两个区间存在重叠但上下限均不同。较宽区间可能同时涵盖充电表现，或来自不同测试集合；报告未作说明。

Treatment: both are quoted at their original locations and never merged into a single figure. Where a single number is required, the 18–22% highway-specific range is used and labelled as such.处理：两处数值均在原位置引用，绝不合并为单一数字。需要单一数值时，使用 18–22% 的高速专项区间并明确标注。

What the source says: owner forums account for 21% of channel share; only 12 verified owner reports are included.原文表述：车主论坛占渠道份额 21%；仅纳入 12 份经核验车主报告。

Issue: 21% of 127 entries is approximately 27 entries, more than twice the 12 verified owner reports. The 21% therefore describes channel sourcing, not verified owner authorship.问题：127 条的 21% 约为 27 条，是 12 份经核验车主报告的两倍以上。因此 21% 描述的是渠道来源，而非经核验的车主作者身份。

Treatment: both figures retained with distinct labels — "channel share" and "verified owner reports". They are never added or substituted for one another.处理：两项数据分别保留并使用不同标签——“渠道份额”与“经核验车主报告”。二者不叠加、不互相替代。

What the source says: 91%, 88%, 87%, 79%, 47.6%, 12.9%, 79.6%, 62%, 3.2%, 12–18%.原文表述：91%、88%、87%、79%、47.6%、12.9%、79.6%、62%、3.2%、12–18%。

Issue: denominators include the hybrid subset (91%), all content (88%), general entries (87%), all entries (79%), all negative feedback (47.6%), all neutral feedback (12.9% and 79.6%), negative owner feedback (62%) and all feedback (3.2%). Cross-comparison of these values is not valid. Percentages also imply non-integer entry counts (22% of 127 = 27.9).问题：分母包括混动子集（91%）、全部内容（88%）、一般性条目（87%）、全部条目（79%）、全部负面反馈（47.6%）、全部中性反馈（12.9% 与 79.6%）、负面车主反馈（62%）以及全部反馈（3.2%）。跨项比较不成立。此外百分比反推出的条目数并非整数（127 的 22% 为 27.9）。

Treatment: each percentage is presented with its denominator stated in the adjacent label or note; the denominator map in Section D makes this explicit for all published values.处理：每个百分比均在相邻标签或注释中标明分母；D 部分的分母对照表对此作出统一说明。

What the source says: owner communities: 30% negative; 62% of negative owner feedback is software-related.原文表述：车主社群负面占比 30%；负面车主反馈中 62% 与软件相关。

Issue: with 12 verified reports, one report moves the negative share by roughly 8 percentage points. The 62% figure corresponds to a very small number of individual complaints. Direction is credible; magnitude is not stable.问题：在 12 份经核验报告下，单份报告即可使负面占比变动约 8 个百分点。62% 对应的是极少数个体投诉。方向可信，幅度不稳定。

Treatment: retained with the sample size stated alongside every owner-side figure. Presented as a signal requiring primary validation, not as a measured population share.处理：保留数值，并在每处车主侧数据旁标注样本量。将其呈现为需一手验证的信号，而非人群实测占比。

What the source says: a bar chart titled "CLA Discussion Volume by Product Phase" with y-axis values 0–40 and a legend labelled "volume_share"; only two of four bars carry labels ("Launch Announcement", "Delivery Start").原文表述：柱状图“CLA Discussion Volume by Product Phase”，纵轴 0–40，图例标注“volume_share”；四根柱中仅两根有标签（“Launch Announcement”“Delivery Start”）。

Issue: the axis carries no unit, so values between 22 and 38 could read as absolute discussion counts. Two phases are unlabelled, and the second bar (38, the peak) has no label at all.问题：坐标轴无单位，22 至 38 之间的数值可能被读作绝对讨论条数；两个阶段缺少标签，且第二根柱（峰值 38）完全没有标签。

Treatment: re-rendered with percent units, all four phases labelled, peak highlighted, and a caption stating these are shares, not counts.处理：重绘为百分比单位，补全四个阶段标签，高亮峰值，并在图注中说明其为占比而非条数。

What the source says: the CLA 250+ is positioned as 12–18% more expensive than the Tesla Model 3 Highland, BMW i4 and Polestar 2 [1].原文表述：CLA 250+ 较 Tesla Model 3 Highland、BMW i4 与 Polestar 2 贵 12–18% [1]。

Issue: no competitor price points, trim levels, market, currency basis or date are provided. The band is a single-source assertion and cannot be audited from the report.问题：未提供竞品价格点、配置级别、市场、货币基准或取数日期。该区间为单一来源的判断，无法依据报告本身审计。

Treatment: displayed as a range band, not as per-competitor bars, so no implied model-level precision. Flagged for re-verification before pricing use.处理：以区间带呈现，而非分竞品柱状，避免暗示车型级精度。已标注在定价使用前需重新核验。

What the source says: negative feedback claims carry [1]; neutral feedback claims carry [7].原文表述：负面反馈结论标注 [1]；中性反馈结论标注 [7]。

Issue: two thematically organised analyses underpin the entire negative and neutral evidence set. No independent corroboration is cited for the 47.6%, 62%, 18–22% or 12.9% figures, and neither reference supplies a URL or access date.问题：整个负面与中性证据集合仅由两份专题分析支撑。47.6%、62%、18–22%、12.9% 等数值均无独立交叉佐证，且两项文献均未提供网址或访问日期。

Treatment: source dependency is stated wherever [1] or [7] carries a load-bearing claim. Findings are presented as the report's claims, not as corroborated facts.处理：凡 [1] 或 [7] 支撑关键结论处，均注明来源依赖。相关发现以“报告声称”呈现，而非“已获佐证的事实”。

What the source says: 91% of hybrid feedback originates from pre-production media test units; all negative hybrid feedback comes exclusively from media test drives; no production owner reports exist.原文表述：91% 的混动反馈来自试装媒体测试车；全部混动负面反馈仅来自媒体试驾；不存在量产车车主报告。

Issue: a calibration judgement formed on pre-production hardware cannot be assumed to describe the delivered vehicle. The report acknowledges this but still ranks hybrid calibration as the largest negative driver.问题：基于试装硬件形成的标定判断，不能假定其描述量产交付车辆。报告承认这一限制，但仍将混动标定列为最大负面动因。

Treatment: retained as the report's finding, always paired with the pre-production qualification. Presented as a launch-readiness signal rather than a market verdict on the production car.处理：作为报告结论保留，但始终附带“试装车”限定；定位为上市准备度信号，而非量产车的市场定论。

What the source says: "The model was awarded European Car of the Year 2026 [2]".原文表述：“该车型荣获 2026 欧洲年度车 [2]”。

Issue: the awarding organisation, category, date and eligible model variants are not stated. "Car of the Year" is used by multiple organisations, so the title alone does not identify the award.问题：未说明授予机构、奖项类别、日期及适用车型版本。“年度车”被多个机构使用，仅凭称号无法确定具体奖项。

Treatment: quoted as a claim attributed to [2], never presented as a verified fact. Verify organiser and category before any external use.处理：作为 [2] 的声称引用，不作为已核验事实呈现。对外使用前须核验授予机构与类别。