Here are the average global temperatures for the contiguous united states over the last 30 years (measured in of). use the table in the image below to answer the 5 questions that come with it.
1. statistical regression can be sensitive to relatively large numbers. to simplify these numbers, recode each year so that 1985 is coded as year 1 and 2014 is coded as year 30.

2. on a separate page, make a time plot of this data (horizontal axis = coded year,
vertical axis = average temperature). do not connect the dots.

3. using technology, find the regression line that best fits this data. report your result in the form
average temp = + (coded year)

4. what average temperature does your regression equation predict for the year 1997?

5. what was the actual average temperature in the year 1997?

1).  The years are already coded that way in the table.

2).  Draw the graph.  It'll have 30 points on it.
The first point is        (1, 51.30) .
The second point is (2, 53.32) .
The third point is       (3, 53.33) .
etc.
For each point, 'x' is the code of the year,
and 'y' is the Avg Temp of that year.

3).  The points don't all exactly fall on a straight line.
You must have learned how to find the "best fit" line.
I used to do a lot of that, but I don't remember the details.
If you crank all 30 points through the linear regression process,
  you get the slope and the y-intercept of the line that fits the data best.
That's what #3 is asking for.  The first blank is the y-intercept of it,
  the second blank is the slope of it.

4).  1997 is code-13.  Plug '13' into your regression equation (#3)
and see what temperature pops out.

5).  From the table, the actual for 1997 was 52.20 .