OK. "Linear model" just means a straight-line graph. That's no problem here, because we know two points on the line, and when we know two points, we know (??) how to find the equation of the line.
The equation is going to be
Y = m t + b .
Y is the chicken consumption in pounds
m is the slope of the line
t is the number of years
b is the y-intercept of the line
One point: 1983. t=3, y=30.9 pounds
Other point, 1989: t=9, y=47 pounds
Slope = (47-30.9) / (9-3)
Slope = 16.1/6 = 2.683 pounds per year
Plug that number into the equation for either of the points that we know, and you find that the y-intercept b is 22.85 pounds. So ...
Y = 2.683 t + 22.85 .
(and t is the number of years after 1980)
In 1995, t=15 . Stick that into the equation, and you'll get the estimate for chicken consumption in 1995, IF your model ... based on changes from 1983 to 1989 ... is still accurate for 6 more years after 1989.
