Qoymans Multiple-Choice #1 statistics

Norms ("G" = raw score)

G Protonorm
0 167 (negative raw scores are possible)
1 173
2 187
3 193
4 207
5 220
6 233
7 247
8 260
9 267
10 287
11 319
12 321
13 326
14 328
15 332
16 334
17 339
18 341
19 346
20 348
21 351
22 352
23 353
24 354
25 355
26 360
27 367
28 378
29 379
30 381
31 382
32 387
33 392
34 394
35 400
36 406
37 408
38 413
39 440
40 446
41 448
42 467
43 493
44 540
45 580
46 587
47 600
48 607
49 620
50 627
51 633
52 647
53 653
54 667
55 680
56 693
57 707
58 727
59 753
60 787

Comment

This test was created to for once do something completely different. I had always seen multiple-choice tests as inferior (and still do) so I made an extreme one. It had too low estimated g loading and too low reliability. Therefore I later combined it with another test of the same type into a larger test and removed or revised bad items. The result is the QMC #4, which is better than the earlier ones, though still the least satisfactory test I have made. The QMC tests have the lowest g loadings of all my tests (so the lowest validity; for an IQ test, validity is equal to g loading). I attribute this entirely to the multiple-choice format.

Norming method

Used for norming are scores from other tests correlating .39 and higher with the QMC #1. The other scores are converted to a scale with SD=15 if needed. Ranks of raw and other scores are equated, and there is extrapolation below raw score 3 and above 45.

Scores (test can no longer be taken)

# testees: 100
Mean: 28.75
SD: 9.92
Male mean: 29.4
Female mean: 25.97 (19 persons)

3*
6*
8*
9**
11*
13***
15***
16*
17***
18*
19**
20*
21*
22****
23***
24*
25***
26*******
27****
28****
29***
30***
31******
32***
33*****
34***
35***
36*****
37***
38*
39****
40***
41**
43***
44.5*
45**
46*
47*
48*

National averages

Correlation with other tests

Estimated g loading: 0.4

Correlation with personal details

Correlation with national IQs

.09 over 35 pairs.

Reliability and standard error of measurement

Rel = 0.83
SE = 3.8 raw score points

Scores by age

Additional statistics March 2007

The below statistics result from entering a number of old scores into the database that had previously been left out because they are from people from whom almost no data is known. The types of statistics that are not given below have not changed compared to the above report.

All scores

• # candidates: 185
• Median score: 26.0
• Maximum possible score: 60
• Mean: 25.4
• SD: 9.88
• Male mean: 26.1
• Female mean: 23.0 (43 persons)
• Hardness: 0.29

Scores by males

n = 142

3*
6**
7**
8*
9*
10***
11**
12***
13****
15****
16**
17*****
18***
19*****
20*
21***
22***
23******
24*****
25*****
26**********
27****
28*********
29******
30***
31********
32***
33******
34***
35****
36****
37***
38**
39****
40****
41**
43**
45**
46*
47*

Scores by females

n = 43

9*
10*
11*
12*
13****
14*
15***
16*
17**
18**
19**
20**
21*
22*****
23**
25**
26*
29*
30*
32*
36**
37*
39*
43*
44.5*
45*
48*

Correlation with other tests

Weighted average of correlations: 0.145

Conservatively estimated minimum g loading: 0.38

Ranking in above table is based on the unrounded correlations. All available data is present in this table, no tests are left out except for those with less than 4 score pairs. All known pairs are used to obtain the true, honest statistics; correlations are not artificially inflated by leaving out ceiling scores, outliers or other anomalies.

Correlation with personal details

Correlation with personal details - within females

Correlation with national IQs

Correlation of this test with national average IQs published by Lynn and Vanhanen:

• r = 0.12
• n = 36