A SOCIAL CRISIS IN STATISTICAL SOCIAL SCIENCE, MEDICAL, AND EDUCATIONAL MATH IN COLLEGE AND UNIVERSITY RESEARCH

Abstract: A number of leading scholars appear to be coming to a similiar conclusion. Social , medical and educational research is conducted in such a way that the results of many “quality” studies produce false positives. Thus, the study appears to valid, but are not. The author suggest returning to traditional measures and refining the new strategies.

Introduction: During the excitement of the beginning of the “Big Data Days” circa mid `70’s, marginal social sciences and related could become more scientific. The author would prefer to use the term “behavorial methodologies” (not science and not art, somewhere in the middle.)

Thus with new software programs and better computers, huge numbers could be organized in such a way for 20th century goodness of fit, correlation and related in a very fast and efficient ways.

Description:

Statistics for soft numbers could be translated into hard number analysis. Thus, a computer printout could deliver not only soft “cross tabs” but “analysis of variance” “multiple regression” “step-wise regression”, and “path analysis.” The regression was all based on hard numbers.

How could that rather dramatic jump in number properties from soft to hard be completed in a short period of time? Thus social scientist wanted the prestige. With their best heads together, the social staticians held sway. Further, graduate students were now math oriented and a number of disciplines could catch up with the arrogant economists.

The author recalls the professor who taught this new ground breaking advanced multi-regression class. He was arrogrant and elitist. Hard numbers is American Pie. From the early elementary days of A,B,C’s to 1-10, everything was hard numbers. Top athletes had hard number ratings. Top sciences that begin with hard numbers were also at the top of the heap. My professor only slipped once when his lipped quivered ad he stated that the course assumed hard numbers.

Further, theoretrically Frued and Skinner were challeged by THIRD FORCE social psychology. Hard number Multiple regression would put the behavorial sciences right up at the top again.

COMING CLEAN:

To add to the crisis, Meta-analysis which is shaky at best was introduced as it was easy to do and had some credence. Graduate students could do all the work. Ioannidis and others were the first to see the sham. All of these wondorous strategies indicated false positives when compared to traditional, less flashy measures.

SOFT NUMBERS:

The author wants to do two major items. The first is to reintroduce soft numbers theory and its value. The second is to give a brief review of all of the criticisms of Ioanndis. The last is to suggest ways in which new and old can compliment each other.

NOMINAL NUMBERS

Years ago, the “social sciences” had to rely on soft numbers. Stat books were filled with statistics that would compliment nominal tests, ordinal tests, and interval statistical tests. Thus a number of fields were held in low esteem because they could not quality for rational analysis assessment.

What the author is leading to is the following. Many areas such as psychology, sociology and political science ofteen worked data that was not hard data. Economics coul d quickly transfer to ratio numbers because a dollar amount was assigned to a concept.

In an attempt to re-honor softer numbers, one can see that statistics to gain prestige is not what search for validity is all about. Rather, an accurate answer with less powerful numbers is still an accurate account. It can compliment ratio numbers perhaps at a later date.

NOMINAL NUMBERS PART TWO

Let’s look at a nomial question. “If the election were held tommorrow, would you vote for John Doe?

yes no . What we gain from this is that the potential voter indicates that they will or will not vote for John Doe. That is quite a lot of information. We do not know the intensisty of the vote or if they will vote. However, we have some indication that they will vote. Now, intensity is one word for a powerful force in the universe. We can get a measurement of increasing strenth, but we will pay dearly for it.

In the end, a vote is a vote. The candidate with the most votes wins the election. Let’s look at two votes for John Doe. The first voter sent One thousand dollars to John Doe and drove through a storn than the second voter. This second person voter did so at the last moment and actually made up their minds a few seconds before voiting. The two John Doe voters are considerably different, but they ended up voting the same way.,A SOCIAL CRISIS IN STATISTICAL SOCIAL SCIENCE, MEDICAL, AND EDUCATIONAL MATH IN COLLEGE AND UNIVERSITY RESEARCH

Abstract: A number of leading scholars appear to be coming to a similiar conclusion. Social , medical and educational research is conducted in such a way that the results of many “quality” studies produce false positives. Thus, the study appears to valid, but are not. The author suggest returning to traditional measures and refining the new strategies.

Introduction: During the excitement of the beginning of the “Big Data Days” circa mid `70’s, marginal social sciences and related could become more scientific. The author would prefer to use the term “behavorial methodologies” (not science and not art, somewhere in the middle.)

Thus with new software programs and better computers, huge numbers could be organized in such a way for 20th century goodness of fit, correlation and related in a very fast and efficient ways.

Description:

Statistics for soft numbers could be translated into hard number analysis. Thus, a computer printout could deliver not only soft “cross tabs” but “analysis of variance” “multiple regression” “step-wise regression”, and “path analysis.” The regression was all based on hard numbers.

How could that rather dramatic jump in number properties from soft to hard be completed in a short period of time? Thus social scientist wanted the prestige. With their best heads together, the social staticians held sway. Further, graduate students were now math oriented and a number of disciplines could catch up with the arrogant economists.

The author recalls the professor who taught this new ground breaking advanced multi-regression class. He was arrogrant and elitist. Hard numbers is American Pie. From the early elementary days of A,B,C’s to 1-10, everything was hard numbers. Top athletes had hard number ratings. Top sciences that begin with hard numbers were also at the top of the heap. My professor only slipped once when his lipped quivered ad he stated that the course assumed hard numbers.

Further, theoretrically Frued and Skinner were challeged by THIRD FORCE social psychology. Hard number Multiple regression would put the behavorial sciences right up at the top again.

COMING CLEAN:

To add to the crisis, Meta-analysis which is shaky at best was introduced as it was easy to do and had some credence. Graduate students could do all the work. Ioannidis and others were the first to see the sham. All of these wondorous strategies indicated false positives when compared to traditional, less flashy measures.

SOFT NUMBERS:

The author wants to do two major items. The first is to reintroduce soft numbers theory and its value. The second is to give a brief review of all of the criticisms of Ioanndis. The last is to suggest ways in which new and old can compliment each other.

NOMINAL NUMBERS

Years ago, the “social sciences” had to rely on soft numbers. Stat books were filled with statistics that would compliment nominal tests, ordinal tests, and interval statistical tests. Thus a number of fields were held in low esteem because they could not quality for rational analysis assessment.

What the author is leading to is the following. Many areas such as psychology, sociology and political science ofteen worked data that was not hard data. Economics coul d quickly transfer to ratio numbers because a dollar amount was assigned to a concept.

In an attempt to re-honor softer numbers, one can see that statistics to gain prestige is not what search for validity is all about. Rather, an accurate answer with less powerful numbers is still an accurate account. It can compliment ratio numbers perhaps at a later date.

NOMINAL NUMBERS PART TWO

Let’s look at a nomial question. “If the election were held tommorrow, would you vote for John Doe?

yes no . What we gain from this is that the potential voter indicates that they will or will not vote for John Doe. That is quite a lot of information. We do not know the intensisty of the vote or if they will vote. However, we have some indication that they will vote. Now, intensity is one word for a powerful force in the universe. We can get a measurement of increasing strenth, but we will pay dearly for it.

In the end, a vote is a vote. The candidate with the most votes wins the election. Let’s look at two votes for John Doe. The first voter sent One thousand dollars to John Doe and drove through a storn than the second voter. This second person voter did so at the last moment and actually made up their minds a few seconds before voiting. The two John Doe voters are considerably different, but they ended up voting the same way.,A SOCIAL CRISIS IN STATISTICAL SOCIAL SCIENCE, MEDICAL, AND EDUCATIONAL MATH IN COLLEGE AND UNIVERSITY RESEARCH

Abstract: A number of leading scholars appear to be coming to a similiar conclusion. Social , medical and educational research is conducted in such a way that the results of many “quality” studies produce false positives. Thus, the study appears to valid, but are not. The author suggest returning to traditional measures and refining the new strategies.

Introduction: During the excitement of the beginning of the “Big Data Days” circa mid `70’s, marginal social sciences and related could become more scientific. The author would prefer to use the term “behavorial methodologies” (not science and not art, somewhere in the middle.)

Thus with new software programs and better computers, huge numbers could be organized in such a way for 20th century goodness of fit, correlation and related in a very fast and efficient ways.

Description:

Statistics for soft numbers could be translated into hard number analysis. Thus, a computer printout could deliver not only soft “cross tabs” but “analysis of variance” “multiple regression” “step-wise regression”, and “path analysis.” The regression was all based on hard numbers.

How could that rather dramatic jump in number properties from soft to hard be completed in a short period of time? Thus social scientist wanted the prestige. With their best heads together, the social staticians held sway. Further, graduate students were now math oriented and a number of disciplines could catch up with the arrogant economists.

The author recalls the professor who taught this new ground breaking advanced multi-regression class. He was arrogrant and elitist. Hard numbers is American Pie. From the early elementary days of A,B,C’s to 1-10, everything was hard numbers. Top athletes had hard number ratings. Top sciences that begin with hard numbers were also at the top of the heap. My professor only slipped once when his lipped quivered ad he stated that the course assumed hard numbers.

Further, theoretrically Frued and Skinner were challeged by THIRD FORCE social psychology. Hard number Multiple regression would put the behavorial sciences right up at the top again.

COMING CLEAN:

To add to the crisis, Meta-analysis which is shaky at best was introduced as it was easy to do and had some credence. Graduate students could do all the work. Ioannidis and others were the first to see the sham. All of these wondorous strategies indicated false positives when compared to traditional, less flashy measures.

SOFT NUMBERS:

The author wants to do two major items. The first is to reintroduce soft numbers theory and its value. The second is to give a brief review of all of the criticisms of Ioanndis. The last is to suggest ways in which new and old can compliment each other.

NOMINAL NUMBERS

Years ago, the “social sciences” had to rely on soft numbers. Stat books were filled with statistics that would compliment nominal tests, ordinal tests, and interval statistical tests. Thus a number of fields were held in low esteem because they could not quality for rational analysis assessment.

What the author is leading to is the following. Many areas such as psychology, sociology and political science ofteen worked data that was not hard data. Economics coul d quickly transfer to ratio numbers because a dollar amount was assigned to a concept.

In an attempt to re-honor softer numbers, one can see that statistics to gain prestige is not what search for validity is all about. Rather, an accurate answer with less powerful numbers is still an accurate account. It can compliment ratio numbers perhaps at a later date.

NOMINAL NUMBERS PART TWO

Let’s look at a nomial question. “If the election were held tommorrow, would you vote for John Doe?

yes no . What we gain from this is that the potential voter indicates that they will or will not vote for John Doe. That is quite a lot of information. We do not know the intensisty of the vote or if they will vote. However, we have some indication that they will vote. Now, intensity is one word for a powerful force in the universe. We can get a measurement of increasing strenth, but we will pay dearly for it.

In the end, a vote is a vote. The candidate with the most votes wins the election. Let’s look at two votes for John Doe. The first voter sent One thousand dollars to John Doe and drove through a storn than the second voter. This second person voter did so at the last moment and actually made up their minds a few seconds before voiting. The two John Doe voters are considerably different, but they ended up voting the same way.,