Each month when the Consumer Price Index comes out, news stories about it almost always also mention what happened to the real weekly take home-pay of an average worker with three dependents. Unfortunately, those take-home pay statistics are one of the most misleading and most widely misused numbers the federal government bothers to calculate.

Ar a congressional hearing last week, Sen. James McClure (R-Idaho) complained to Charles Schultze, chairman of the Council of Economic Advisers, that those numbers showed real take home-pay dropping 3 percent last year. And because of the way inflation pushes people into higher tax brackets, "It's obvious that the real spendable income of the average American will continue to decline," McClure argued.

Actually this "spendable earnings series," as these numbers are known, is so misleading that nothing it ever shows is "obvious" and certainly not that most Americans are failing to keep up with inflation. Besides, some much better indicators are available, including real disposable income per person.

Look what can happen to the spendable earnings numbers, for instance when 18-year-old George Anderson takes a part-time job at a new fast-food restaurant opens near his home.

Chances are that he will get the federal minimum wage, which rose last month from $2.60 an hour to $2.95 for the 12 hours he works. His before-tax pay each week will be $35.40.

Anderson will be counted by the Bureau of Labor Statistics as working in a service industry, in which spendable average weekly earnings in December were $153.47. Even though he is only working part time, his pay gets aversged in with that of all the workers in the industry, including those working full time.

So Anderson's going to work in a new job will pull down the averages, both because of his low wage and part-time status. But does that mean that somehow an average American worker is worse off? Hardly.

On a vastly larger scale, that's exactly what has been happening to the spendable earnings numbers in recent years. "The entry of women and teenagers into the labor force in such large numbers has made the earnings series rise very little," according to Marvin Kosters, a labor economist at the American Enterprise Institute.

The spendable earnings figures for the total private economy -- the numbers cited most often -- also are affected by shifts in the relative numbers of employes in high-paying and low-paying industries. Employment has grown relatively little in mining and manufacturing, but very rapidly in wholesale and retail trade, and in services. The former pay far better, so that this change, too, has held down the averages.

Nor, despite McClure's attempt to use the series to make a point about taxes, does the series really reflect tax burdens adequately, most economists say.

The spendable earnings figures are adjusted for rederal personal income tax withholding and for Social Security. State and local taxes are not counted. And even Social Security tax changes may not be counted fully.

Solely because of last year's $12 billion personal income tax cut, average weekly take-home pay will jump about $1.75 in January for that worker with three dependents who was betting $187.23 in December. On the other hand, last month's small Social Security tax rate increase -- from 6.05 to 6.13 percent -- will trim about 15 cents.

But the big Social Security tax increase this year is in the so-called wage base -- the amount of one's earnings on which the tax is due. That jumps from $17,700 last year to $22,900 for 1979. However, because none of the "average" workers depicted in the spendable earnings series makes enough to be affected by that change, it will not show up at all.

Two other factors that can influence the weekly- take-home-pay figures are shifts in the length of the average work week and changes in the amount of overtime worked.

When the amount of overtime goes up or down, it pushes take-home pay in the same direction. More importantly, the Council of Economic Advisers estimates that the length of the average work week is declining by about one-half of one percent each year. For that reason alone, spendable earnings are rising that much less than they otherwise would be.

All in all, this particular statistical series is so misleading that "people don't want to try to make it better, they want to throw it out and put something else in its place," said one BLS economist. And he added, "It has some validity, but not for what it's used for."

Compare what has happened to real spendable earnings for a married worker with three dependents in the last 10 years, on the one hand, with what has happened to real disposable personal income per person, on the other.

In 1968, the real spendable earnings for that worker stood at $91.44. By last year, if you believed the numbers, his real take-home pay had risen to only $92.49. That's an annual increase of only about one percent.

But look at real disposable personal income per person. It includes the full effect of Social Security and all other federal taxes, not just the personal income tax. Moreover, it is calculated after state and local tax payments also are deducted.

By this measure, real incomes have gone up 27.5 percent over the last 10 years despite all the federal, state and local tax increases and years of high inflation.

Here, too, numbers can be misleading, cautioned AEI's Kosters. Part of this 27.5 percent increase is due to the ever-growing proportion of the nation's population that is working. Last month, a record 59.3 percent of all noninstitutionalized, working-age Americans were at work.

At least when George Anderson takes his part-time job, this set of statistics recognizes that the nation's income is higher, not lower. So the next time you read that the real take-home pay of an average married worker with three dependents went up or down, take it with a grain of salt -- or less.