+ United States Explore a state. Follow the impact.
Representatives
THE PATTERN LABCOMPARE → QUESTION → VERIFY
Does the pattern hold together?
Put two measures beside each other. Follow a state through time. See the relationship—and the
records behind it.
01 / CHOOSE A YEAR2025
December 2024 → December 2025Same calendar month. Changes, not raw levels.
Two establishment-survey measures
Private-sector payroll jobs and average hourly earnings; both not seasonally adjusted. Earnings are nominal, not adjusted for inflation.
02 / EXPLORE THE RELATIONSHIP
↑ Average hourly earnings year-over-year % change · nominal dollars
-0.91.43.65.98.1
-3.9-2.3-0.80.72.3
Private payroll jobs → year-over-year % change
TexasSelected state0.62%
private payroll jobs · 3.25% hourly earnings
Read a dot Hover or focus to trace its two values. Click or press Enter to select. Keyboard: ← / → browse by horizontal value; ↑ / ↓ by earnings. Home / End browse
alphabetically. Tab leaves the chart.
Both increasedPay ↑ / other ↓Pay ↓ / other ↑Both decreasedAt least one unchanged
Colors describe directions, not “good” or “bad.” Axes fit this year; visual distances are not comparable across years. Map areas without a
complete pair remain uncolored.
PEARSON r0.10Descriptive linear association
51 region pairs. One year.
One dot per state or DC; every region gets equal weight. 0 excluded for
missing endpoints or an unusable baseline. Selecting a state does not change this calculation.
r runs from −1 to +1. Values near zero mean little linear association—not “no
relationship.” This is not a causal effect, forecast, or significance test.
BEHIND THE COLORSWhere did the changes land?51 regions +
December 2024 → December 2025. The strip counts all 51
collected regions, including incomplete pairs. Open a group, then select a region to mark it on
the chart and map.
Pay increased / jobs increased21⌄
Pay increased / jobs decreased30⌄
Pay decreased / jobs increased0⌄
No regions in this group for this year.
Pay decreased / jobs decreased0⌄
No regions in this group for this year.
At least one measure unchanged0⌄
No regions in this group for this year.
No complete pair0⌄
No regions in this group for this year.
Groups describe changes relative to zero, not a ranking or a strength score. They do not
filter the chart or recalculate r. Direction uses unrounded values; very small nonzero changes
are marked explicitly. Pay means nominal average hourly earnings.
ZOOM OUT / 9 SEPARATE YEARS
Does the relationship persist?
r −1 to +1
One year can tell a different story. Each point recalculates the correlation across that year’s
complete region pairs.
+10−1
2025 · r +0.10
201720182019202020212022202320242025
2025Selected year
r +0.1051 included · 0 excluded
The same 51 regions are included in every shown year. Selecting
a state never changes these annual samples.
Hover or use ← / → to preview; click or press Enter to open that year’s map and records. Home /
End jump to the first / last year. These are separate annual estimates, not pooled years or
evidence of cause.
Read every year’s values and coverage
Pay + jobs · December-to-December
changes · unweighted region pairs
Year
Pearson r
Included
Excluded
+0.16
51
0
+0.07
51
0
+0.15
51
0
-0.30
51
0
-0.19
51
0
+0.23
51
0
+0.05
51
0
+0.04
51
0
+0.10
51
0
03 / FOLLOW A STATE
TWO REGIONS / THE SAME WINDOW
Put the changes side by side.
Choose another region to compare its changes with Texas. The chart’s correlation
continues to use all complete region pairs.
THE SAME REGION · TWO MEASURES
Texas
Private payroll jobs+0.62%12208.7 → 12284.7
thousand private payroll jobs
Average hourly earnings+3.25%$34.16 → $35.27 per hour · nominal
Average earnings can rise because the mix of jobs changed. This does not measure the raise
received by a typical worker.
For each region, pay and job changes are (December value ÷ previous December value − 1) ×
100. Unemployment change is the difference between the two rates in percentage points.
Pearson r uses only complete pairs for the selected year, with no population weighting or
imputation. At least three pairs and variation on both axes are required.
These aggregates are not independent experimental observations. Shared economic shocks,
commuting, job mix and regional differences may affect the pattern. LAUS also uses payroll
information among its model inputs, so the programs are not wholly independent. We do not
calculate a p-value or infer an individual-level relationship.