<table cellspacing="0" cellpadding="0" border="0" ><tr><td valign="top" style="font: inherit;"><DIV>Dear Sir/Madam,</DIV>
<DIV> </DIV>
<DIV>I am writing to know whether the t-score used in corpus<BR>analysis is the same t-score used in regular statistics. That is, if I am,<BR>for example, looking for the collocation ‘wing’ and ‘angel,’ and I find<BR>that these two words occur together 75 times with a t-score value of 4.3,<BR>can I say that the df (<SPAN class=yshortcuts id=lw_1238588039_0 style="CURSOR: hand; BORDER-BOTTOM: #0066cc 1px dashed">degree of freedom</SPAN>) is 75-1=74, and then go to the<BR>t-score table and try to find whether my result is statistically<BR>significant, i.e. p<0.05?</DIV>
<DIV> </DIV>
<DIV>I look forward to your response.</DIV>
<DIV> </DIV>
<DIV>Sincerely,</DIV>
<DIV> </DIV>
<DIV>Sandy<BR></DIV></td></tr></table><br>