A study into germination of parsnip seeds in the 1980s produced the following data for the average number of days, \(d\), taken for a seed to germinate at different soil temperatures, \(t\), measured in degrees Fahrenheit.
| \(t\) | \(32\) | \(41\) | \(50\) | \(59\) | \(68\) |
| \(d\) | \(172\) | \(57\) | \(27\) | \(19\) | \(14\) |
(i) Plot \((32,172),(41,57),(50,27),(59,19),(68,14)\), with \(t\) on the horizontal axis and \(d\) on the vertical axis. The correlation is \(r_{dt}=-0.848\).
(ii) With \(u=1/t\), least-squares regression gives \(d=-145.12547+9456.40257u\). Thus \(a=145\) and \(b=9.46\times10^3\) to 3 significant figures, and \(r_{du}=0.941\).
(iii) At \(t=86\), the model predicts about \(-35.2\) days, which is impossible and far from the observed \(32\). It does not fit the additional data.
(iv) Since \(F=\frac95T+32\), \(d=-145+\frac{9460}{\frac95T+32}\), using the rounded coefficients from part (ii).
Need help? Join our JC Math tuition classes.
Learn more