ICE — Excel Functions (SLOPE/INTERCEPT/RSQ/STEYX)

Prefer the one-click Excel summary? 👉 Use the ToolPak version. Both use the same dataset

Step A — Use this dataset (Sleep → Exam)

Copy to Excel: click inside A1 and paste. Headers included. If columns stick together, use Data → Text to Columns → Delimited → Tab.

#	X (Sleep hours)	Y (Exam score)

Step B — One-cell Excel formulas (click to copy)

Assumes your data are in A1:C24 with headers on row 1 (X in A2:A24, Y in B2:B24). Paste each into an empty cell.

Slope (b1)

Intercept (b0)

R²

r (correlation)

STEYX (typical miss)

Forecast at x₀ in H2

FORECAST.LINEAR

Residuals (in C2, fill down)

Optional (t/p/ANOVA) via LINEST — copy to any single cell

Returns the full LINEST table (5×2). Use INDEX to pick out cells, or select a 5×2 range and enter as dynamic array.

LINEST block

t(b1) from LINEST

df (n−2)

p(two-tailed) for b1

For a full ANOVA table instantly, use the ToolPak version (Data → Data Analysis → Regression).

Step C — What to write (plain English)

Slope: “+1 hour of sleep is associated with ≈ b1 points higher exam score (units: points/hour).”
Intercept: expected Y at X=0 (if 0 hours isn’t in your data, treat as algebra only).
Fit: “R² = value ⇒ about R²×100% of variation is explained; typical vertical miss ≈ STEYX points.”
Forecast: pick x₀ inside your X-range; report ŷ and a rough band ±2·STEYX. Avoid extrapolation.
Checks: residuals vs X ~ random cloud; no curve or huge outliers.

Good model cues: higher R², smaller STEYX (in Y-units), tiny p for slope (if you used LINEST), and “messy-random” residuals.

Quick Write-Up Template

Paste this into your report and replace the numbers with your Excel results.

Problem. Predict Exam score (Y, points) from Sleep hours (X, hours) with a straight line.

Model (Excel). Using =SLOPE, =INTERCEPT, =RSQ, =CORREL, =STEYX:
ŷ = [b0] + [b1]·X.
Fit. R² = [R2] (≈[xx]% explained); r = [r]; STEYX = [steyx] points.
Forecast. At X = [x0] hours, ŷ = [yhat] points (≈ ŷ ± 2·STEYX).
Interpretation. +1 hour → ≈ [b1] points change in Y. Intercept at X=0 is [in-range/out-of-range].
Checks. Residuals show no obvious curve/outliers; no extrapolation in forecast.

Need the instant ANOVA & p-values?

Jump to the ToolPak path (Excel’s Data → Data Analysis → Regression) with the same dataset and a student-friendly legend:

Open the ToolPak ICE

ICE — Excel Functions Path (No ToolPak)

Step A — Use this dataset (Sleep → Exam)

Step B — One-cell Excel formulas (click to copy)

Step C — What to write (plain English)

Quick Write-Up Template

Need the instant ANOVA & p-values?