Reading Conflicting Reports on a Provincial Crop Failure
A news desk receives three reports on the same province's wheat harvest. Report 1, from a ministry press release, says production rose 8 percent. Report 2, from a local farmers' association, says the harvest fell by a third. Report 3, from an aid agency, says rainfall was 40 percent below average and that 12,000 households need assistance, but gives no production figure. The journalist has one day before publication. She learns that the ministry figure covers the whole province while the farmers' figure covers one district; that the ministry changed its baseline this year; and that the aid agency's rainfall data comes from a single weather station.
1. Assess the credibility and comparability of each report and identify what can and cannot be concluded.
Read the worked answer
Report 1: an 8 percent rise is not comparable to the farmers' claim because the ministry changed its baseline - a rise against a revised baseline may still be a fall against the old one. Also, a province-wide figure can mask a severe district-level drop.
Report 2: a one-third fall is district-level, so it cannot be set against the province-wide figure. It is credible as a local observation but not generalizable.
Report 3: the 40 percent rainfall deficit rests on one station, so it is weak evidence for the whole province, but the 12,000 households needing assistance is a direct observation of need and does not depend on the rainfall estimate.
What can be concluded: the three reports are not measuring the same thing (different areas, different baselines, different indicators), so the apparent contradiction is largely an artifact of comparison. What cannot be concluded: whether provincial production rose or fell, and whether the district's decline is typical. The aid agency's caseload is the most directly usable fact.
2. Draft the core of a fair news paragraph and justify each choice of wording.
Read the worked answer
Model paragraph: 'Official figures show provincial wheat production up 8 percent, but the ministry changed its baseline this year, so the figure cannot be compared directly with earlier years. In one district, farmers report a harvest about a third smaller than last year; the province-wide figure may conceal such local drops. An aid agency reports rainfall 40 percent below average at one weather station and says 12,000 households need assistance. The agency did not provide a production estimate.'
Justification: the paragraph attributes each figure to its source; flags the baseline change because it affects interpretation; states the district scope of the farmers' figure so readers do not generalize it; notes the single-station limitation on the rainfall number; and separates the caseload, which is a direct observation, from the rainfall inference. No figure is presented as the 'true' harvest because none of the three supports that claim.
3. The editor wants a single headline number. What should the journalist recommend, and why?
Read the worked answer
Recommend against a single headline number. None of the three sources measures provincial production on a consistent basis: the ministry figure uses a changed baseline, the farmers' figure is district-level, and the aid agency gives no production figure. Publishing one number as 'the harvest' would assert a comparison the evidence does not support.
A defensible alternative: lead with the human dimension that is best supported - the 12,000 households reported by the aid agency - and present the production dispute as a dispute, with each figure attributed and its scope stated. If a number must appear in the headline, use the one with the clearest scope and note the caveat in the subheading. Explain to the editor that a precise-sounding but unsupported number would be harder to correct later than an honest statement that the figures conflict.