A “good” R-value depends entirely on what you’re measuring. Most people searching this question want to know one of two things: how much insulation their home needs, or how to interpret a correlation coefficient in statistics. Both use the term “R-value,” but they measure completely different things. Here’s what counts as “good” in each context.
R-Value in Home Insulation
In insulation, the R-value measures how well a material resists heat flow. A higher number means better insulating power. What counts as “good” depends on where you live and which part of your home you’re insulating.
The U.S. is divided into climate zones numbered 1 (hottest, like southern Florida) through 8 (coldest, like northern Alaska). ENERGY STAR recommends these R-values for attic insulation in existing homes:
- Zone 1: R-30
- Zones 2 and 3: R-49
- Zones 4 through 8: R-60
Attics need the highest R-values because heat rises and escapes through the roof. Floors need less: R-25 to R-49 if you’re starting from zero insulation, or R-13 to R-38 if you already have a few inches in place. Walls typically get R-5 to R-10 of insulative sheathing added when siding is replaced, since the wall cavity limits how much material you can fit.
Basement and crawlspace walls follow their own scale. In milder climates (Zone 3), R-5 sheathing or R-13 batts are sufficient. In colder regions (Zones 5 through 8), you’ll want R-15 sheathing or R-19 batts.
What the Numbers Mean in Practice
Different insulation materials deliver different R-values per inch of thickness. Fiberglass batts provide roughly R-3.0 to R-3.8 per inch. Blown-in cellulose lands around R-3.2 to R-3.8 per inch. Closed-cell spray foam is the most efficient, offering about R-6.0 to R-7.0 per inch. So reaching R-60 in your attic might take about 16 inches of fiberglass but only 9 or 10 inches of spray foam. The right choice depends on your budget, your space, and what’s already installed.
R-Value in Statistics (Correlation Coefficient)
In statistics, the Pearson correlation coefficient (r) measures the strength and direction of a straight-line relationship between two variables. It ranges from -1.0 to +1.0. A value of +1.0 means a perfect positive relationship (as one goes up, the other goes up in lockstep), while -1.0 means a perfect negative relationship. Zero means no linear relationship at all.
A widely used rule of thumb breaks it down like this:
- 0.90 to 1.00: Very high correlation
- 0.70 to 0.90: High correlation
- 0.50 to 0.70: Moderate correlation
- 0.30 to 0.50: Low correlation
- 0.00 to 0.30: Negligible correlation
The same thresholds apply to negative values. An r of -0.85 is just as strong as +0.85; it simply means the variables move in opposite directions.
What Counts as “Good” Depends on Your Field
There’s no single number that qualifies as a good correlation across all disciplines. In physics and engineering, researchers typically expect r values above 0.95 before considering a relationship meaningful. The systems they study are tightly controlled, so anything lower suggests the model is missing something important.
In the social sciences, an r above 0.60 is generally considered meaningful. Human behavior is messy and influenced by dozens of unmeasured factors, so correlations are naturally weaker. Cohen’s widely cited guidelines label r = 0.10 as a small effect, r = 0.30 as medium, and r = 0.50 as large for research on individual differences. A “large” effect in psychology would barely register in a physics experiment.
R-Squared: A Related but Different Measure
If you’re working with regression rather than simple correlation, you may be looking at R-squared (R²) instead. R-squared is just the correlation coefficient squared, and it tells you the percentage of variation in one variable that’s explained by the other. An r of 0.70 gives an R² of 0.49, meaning the model explains 49% of the variation.
The benchmarks for a “good” R² vary dramatically by field. Physics and chemistry generally expect values above 0.70. Finance considers 0.40 to 0.70 acceptable. Ecology accepts 0.20 to 0.50 as reasonable, given the complexity of natural systems. In clinical medicine, where human biology and psychosocial factors create enormous variability, an R² above 0.15 (15%) is often considered meaningful.
One important caveat: R² tells you how well a model fits your data, not whether the model is correct. A high R² with the wrong variables is still a bad model. And a low R² doesn’t necessarily mean the relationship is unimportant, it may just mean other unmeasured factors play a large role.
R-Value in Epidemiology
You may also encounter “R-value” in the context of infectious disease, especially after COVID-19 made the term common in news coverage. Here, R₀ (pronounced “R-naught”) represents the basic reproduction number: how many people, on average, one infected person will pass the disease to in a population with no immunity.
The critical threshold is 1.0. An R₀ above 1 means each infected person spreads the disease to more than one other person, so the outbreak grows. Below 1, the outbreak shrinks and eventually dies out. The goal of vaccination campaigns is to push the effective reproduction number below 1.
Different diseases have very different R₀ values. Measles is one of the most contagious diseases known, with an R₀ of 12 to 18, meaning one person typically infects 12 to 18 others. That’s why measles requires 92 to 94% of the population to be immune before herd immunity kicks in. By comparison, highly pathogenic avian influenza and Ebola both have R₀ values below 2. COVID-19 was estimated at 2.5 to 5, depending on the variant and setting. Seasonal influenza sits at roughly 1.4 to 4.
In this context, a “good” R-value is a low one. Public health officials aim to keep the real-time reproduction number below 1 through vaccination, quarantine, and other measures that reduce transmission.