datatalking · September 2, 2016 08:33 · Nov 17, 2009 · Nov 17, 2009 · Nov 17, 2009
diff --git a/mbti.py b/mbti.py
@@ -25,9 +25,8 @@
 def is_significant(prob): return prob < (1.0 - CONFIDENCE_INTERVAL)
 
 print "  Using a binomial test at a %.0f%c confidence level." % (100*CONFIDENCE_INTERVAL, '%')
-
-
-# Compute binomial probabilities
+print ''
+print 'Single-group results:'
 for group in data:
     prob = binom.pmf(group['count'], n, group['expected'])
     freq = 100.0 * float(group['count']) / float(n)
@@ -42,3 +41,60 @@ def is_significant(prob): return prob < (1.0 - CONFIDENCE_INTERVAL)
             freq, '%',
             group['count'], n,
             (100.0 * group['expected']), '%'))
+
+
+print ''
+print 'Pair-wise results:'
+try:
+    from itertools import combinations
+except ImportError:
+    def combinations(iterable, r):
+        # combinations('ABCD', 2) --> AB AC AD BC BD CD
+        # combinations(range(4), 3) --> 012 013 023 123
+        pool = tuple(iterable)
+        n = len(pool)
+        if r > n:
+            return
+        indices = range(r)
+        yield tuple(pool[i] for i in indices)
+        while True:
+            for i in reversed(range(r)):
+                if indices[i] != i + n - r:
+                    break
+            else:
+                return
+            indices[i] += 1
+            for j in range(i+1, r):
+                indices[j] = indices[j-1] + 1
+            yield tuple(pool[i] for i in indices)
+
+pairs = {}
+for pair in combinations(range(4), 2):
+    firstindex = pair[0]
+    secondindex = pair[1]
+
+    for group in data:
+        pairname = ''.join([group['name'][firstindex],
+                            group['name'][secondindex]])
+
+        if not pairname in pairs: pairs[pairname] = {'actual': 0, 'expected': 0}
+        pairs[pairname]['actual']   += group['count']
+        pairs[pairname]['expected'] += group['expected']
+
+
+
+for pairname, pair in pairs.iteritems():
+    prob = binom.pmf(pair['actual'], n, pair['expected'])
+    freq = 100.0 * float(pair['actual']) / float(n)
+
+    if is_significant(prob):
+        sig = 'significant:    '
+    else:
+        sig = 'NOT significant:'
+
+    print ("  %s is %s %5.2f%c (%3d/%d) vs. %5.2f%c expected." %
+           (pairname, sig,
+            freq, '%',
+            pair['actual'], n,
+            (100.0 * pair['expected']), '%'))
+
diff --git a/mbti.py b/mbti.py
@@ -26,13 +26,11 @@ def is_significant(prob): return prob < (1.0 - CONFIDENCE_INTERVAL)
 
 print "  Using a binomial test at a %.0f%c confidence level." % (100*CONFIDENCE_INTERVAL, '%')
 
-sum_of_freqs = 0.0
 
 # Compute binomial probabilities
 for group in data:
     prob = binom.pmf(group['count'], n, group['expected'])
     freq = 100.0 * float(group['count']) / float(n)
-    sum_of_freqs += freq
 
     if is_significant(prob):
         sig = 'significant:    '

diff --git a/mbti.py b/mbti.py
@@ -0,0 +1,46 @@
+from scipy.stats.distributions import binom
+
+data = [
+    {'name': 'ISTJ', 'count': 12,  'expected': 0.1160}, 
+    {'name': 'ISFJ', 'count': 4,   'expected': 0.1380}, 
+    {'name': 'INFJ', 'count': 14,  'expected': 0.0150}, 
+    {'name': 'INTJ', 'count': 206, 'expected': 0.0210}, 
+    {'name': 'ISTP', 'count': 16,  'expected': 0.0540}, 
+    {'name': 'ISFP', 'count': 1,   'expected': 0.0880}, 
+    {'name': 'INFP', 'count': 56,  'expected': 0.0430}, 
+    {'name': 'INTP', 'count': 191, 'expected': 0.0430}, 
+    {'name': 'ESTP', 'count': 3,   'expected': 0.0430}, 
+    {'name': 'ESFP', 'count': 1,   'expected': 0.0850}, 
+    {'name': 'ENFP', 'count': 22,  'expected': 0.0810}, 
+    {'name': 'ENTP', 'count': 56,  'expected': 0.0330}, 
+    {'name': 'ESTJ', 'count': 9,   'expected': 0.0870}, 
+    {'name': 'ESFJ', 'count': 5,   'expected': 0.1230}, 
+    {'name': 'ENFJ', 'count': 15,  'expected': 0.0240}, 
+    {'name': 'ENTJ', 'count': 49,  'expected': 0.0180}, 
+]
+
+n = sum(group['count'] for group in data)
+
+CONFIDENCE_INTERVAL = 0.99
+def is_significant(prob): return prob < (1.0 - CONFIDENCE_INTERVAL)
+
+print "  Using a binomial test at a %.0f%c confidence level." % (100*CONFIDENCE_INTERVAL, '%')
+
+sum_of_freqs = 0.0
+
+# Compute binomial probabilities
+for group in data:
+    prob = binom.pmf(group['count'], n, group['expected'])
+    freq = 100.0 * float(group['count']) / float(n)
+    sum_of_freqs += freq
+
+    if is_significant(prob):
+        sig = 'significant:    '
+    else:
+        sig = 'NOT significant:'
+
+    print ("  %s is %s %5.2f%c (%3d/%d) vs. %5.2f%c expected." %
+           (group['name'], sig,
+            freq, '%',
+            group['count'], n,
+            (100.0 * group['expected']), '%'))
No results found