---

So what are we listening for? A comparison of the English listening constructs
in the Japanese National Centre Test and TOEFL^® iBT

---

Appendix B: Raw data and descriptive statistics
for the 2006 J-NCT examinees used in this study

Table adapted from Henning (1987)
NOTE: "1" indicates a correct response and "O" an incorrect response.
*Items should be rejected if the proportion correct is < .33 or p> .67 (Henning, 1987 citing Tuckman, 1978)

[ p. 88 ]

Item Number	Sum for High Group (n=14)	Sum for Low Group (n=14)	Discrimination Index
1	13	3	.81
2	14	14	.50
3	14	13	.51
4	14	8	.64
5	14	11	.56
6	14	7	.67
7	14	6	.68
8	14	5	.73
9	14	4	.73
10	14	8	.61
11	12	1	.92
12	14	13	.51
13	7	6	.54
14	11	5	.69
15	12	10	.55
16	14	9	.60
17	14	12	.54
18	14	11	.56
19	14	9	.60
20	14	9	.60
21	13	0	1.0
22	12	6	.67
23	11	3	.79
24	13	9	.59
25	10	1	.90

Discrimination indices for the 2006 English Listening Section of the
Japanese National Centre Test when administered to a group of 51 high school students

NOTE: The upper and lower 28% of the total group was used in this study, as recommended by Henning (1987).


According to the discriminability continuum, the acceptable range starts 2/3 of the way along.
Hence if the discrimination index is below .67, the item should be rejected (Henning, 1987).

Main Article	Appendix A	Appendix B	Appendix C	Appendix D	Appendix E
Appendix F	Appendix G	Appendix H	Appendix I	Appendix J	Appendix K

---

2006 Pan SIG-Proceedings: Topic Index Author Index Page Index Title Index Main Index
Complete Pan SIG-Proceedings: Topic Index Author Index Page Index Title Index Main Index

---

[ p. 89 ]