Printer Friendly

Biserial correlation coefficient. Applicative program.

Considerand ca variabila dihotomica poate lua valorile A si B, formula de calcul al coeficientului de corelatie biserial este:

r = [[bar.x].sub.1] --[[bar.x].sub.2]/s * pq/y

unde:

[[bar.x].sub.1] = media valorilor variabilei continue pentru subiectii care au caracteristica A a variabilei dihotomice;

[[bar.x].sub.2] = media valorilor variabilei continue pentru subiectii care au caracteristica B a variabilei dihotomice;

s = abaterea standard a variabilei continue pe intreg lotul de subiecti;

pq/y = valoare care se ia dintr-un tabel special in functie de p, sau de q.

procedure coef_corel_biserial(pq_y:real;var r:real);
   type vect=array[1..50] of real;
   var v,n1,n2:vect;
      n:integer;
      media1,media2,med_tot,s:real;
   procedure citire(k:integer;var x:vect);
      var i:integer;
      begin
      for i:=1 to k do
      begin
      write('Varianta',i:3,' : ');
      readln(x[i]);
      end;
   end;
function media(k:integer;v,x:vect):real;
   var nt,suma:real;
      i:integer;
   begin
      suma:=0;
      nt:=0;
      for i:=1 to k do
      begin
      suma:= suma+v[i]*x[i];
      nt:=nt+x[i];
      end;
      media:=suma/nt;
   end;
function media_total(k:integer;v,x,y:vect):real;
   var nt,suma:real;
      i:integer;
   begin
      suma:=0;
      nt:=0;
      for i:=1 to k do
      begin
      suma:= suma+v[i]*(x[i]+y[i]);
      nt:=nt+x[i]+y[i];
      end;
      media_total:=suma/nt;
   end;
function ab_med_patrat(k:integer;med_total:real;v,x,y:vect):real;
   var nt,suma:real;
      i:integer;
   begin
      suma:=0;
      nt:=0;
      for i:=1 to k do
      begin
      suma:=suma+sqr(v[i]-med_total)*(x[i]+y[i]);
      nt:=nt+x[i]+y[i];
      end;
      ab_med_patrat:=sqrt(suma/nt);
   end;
begin {procedure coef_corel_biserial}
   write('Introduceti numarul de variante al variabilei continue n=');
      readln(n);
      writeln('Introduceti variantele variabilei continue');
      citire(n,v);
      writeln('Introduceti numerele corespunzatoare variantelor care
      poarta caracteristica A a variantei dihotomice');
      citire(n,n1);
      writeln('Introduceti numerele corespunzatoare variantelor care
      poarta caracteristica B a variantei dihotomice');
      citire(n,n2);
      write('Introduceti valoarea pq/y = ');
      readln(pq_y);
      media1:=media(n,v,n1);
      media2:=media(n,v,n2);
      med_tot:=media_total(n,v,n1,n2);
      s:=ab_med_patrat(n,med_tot,v,n1,n2);
      r:=(media1-media2)/s*pq_y;
   writeln('Coeficientul de corelatie biserial r = ',r:6:4);
end;

Bibliografie

1. Costescu, Mihai-Radu, Statistic Methods Applied in Social Sciences, Liberty Press and Publishing House, Panciova, 2007

2. Costescu, Mihai-Radu, Costel, Ionascu, Electronic Information Processing, Universitaria Press, Craiova, 2001

COPYRIGHT 2009 University of Craiova
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2009 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:METODE STATISTICE
Author:Costescu, Mihai Radu
Publication:Revista de Stiinte Politice
Article Type:Report
Date:Jul 1, 2009
Words:430
Previous Article:Unificarea comunitara a dispozitiilor aplicabile in materia consimtamantului actului juridic civil. Analiza comparativa a dreptului continental si a...
Next Article:Jurnalistul roman: deziderate si interpretari.

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters