Hypotheses presence and acceptance in seven soil science journals

• Less than 6% of the soil science papers have a hypothesis.
• Soil science papers have a relatively low acceptance rate of tested hypothesis compared to other scientific disciplines.
• No temporal trends have been detected for the surveyed seven soil journals.

The presence and acceptance rates of hypotheses of papers in seven major soil journals were analyzed between 2001 and 2013. The aim of the study was to quantify the testing of hypotheses in soil science and investigate how it evolved over time. The journals were Applied Soil Ecology, Biology and Fertility of Soils, European Journal of Soil Science, Geoderma, Plant and Soil, Soil Biology and Biochemistry and Soil & Tillage Research. In total 15,344 papers were published by the seven journals over that period. Of a sample of 620 papers, 74% tested one hypothesis, 20% tested two or more hypotheses and 6% proposed a hypothesis. In total 66% of the all tested hypotheses (n = 783) were accepted, and the acceptance rate for the seven journals was more or less constant over time. A single hypothesis is more likely to be accepted (75%) compared to research with multiple hypotheses (55%). Although there was some difference between journals, it was concluded that acceptance rates of hypotheses in soil science are relatively low compared to other scientific disciplines.