Table 22 Sorted observed data matrix - Guttman scalogram |
(controlled by LINELENGTH=).
Table 22.1 Sorted observed data matrix - Guttman scalogram
GUTTMAN SCALOGRAM OF RESPONSES: KID |ACT |1111112 1 221 121 22 |8920311251427634569784035 |------------------------- 2 +2222222222222222222222222 M ROSSNER, LAWRENCE F. 41 +2222222222222222222222212 F FATALE, NATASHA 34 +2222222222222222222222211 F PASTER, RUTH 17 +2222222222222222222222210 F SCHATTNER, GAIL 50 +2222222222222222222122111 M DOEPPNER, TOM 45 +2222222222222222222221200 F MCLOUGHLIN, BILLY 7 +2222222222222222220220220 M WRIGHT, BENJAMIN 16 +2222222222222222222111101 F BUFF, MARGE BABY 48 +2222222222222222212221100 M CHAZELLE, BERNIE 25 +2222222222222222212021101 F SEILER, KAREN 59 +2222222222222222212111020 F CLAPP, LB 18 +2222222222222222212102100 M ERNST, RICHARD MAX |
The observations are printed in order of person and item measures as a Guttman Scalogram, with most able persons listed first, the easiest items printed on the left. This scalogram shows the extent to which a Guttman pattern is approximated. See also Guttman Coefficient of Reproducibility. |
Table 22.2 Guttman scalogram of zoned responses
GUTTMAN SCALOGRAM OF ZONED RES KID |ACT |1111112 1 221 121 22 |8920311251427634569784035 |------------------------- 2 +2222222222222222222222222 M ROSSNER, LAWRENCE F. 41 +22222222222222222222222A2 F FATALE, NATASHA 34 +22222222222222222222222AA F PASTER, RUTH 17 +22222222222222222222222A@ F SCHATTNER, GAIL 50 +2222222222222222222A22A11 M DOEPPNER, TOM 45 +222222222222222222222AB@@ F MCLOUGHLIN, BILLY 7 +222222222222222222@2B@BB@ M WRIGHT, BENJAMIN 16 +2222222222222222222A111@1 F BUFF, MARGE BABY 48 +22222222222222222A22B11@@ M CHAZELLE, BERNIE 25 +22222222222222222A2@B11@1 F SEILER, KAREN 59 +22222222222222222A2A11@B@ F CLAPP, LB 18 +22222222222222222A21@B1@@ M ERNST, RICHARD MAX |
The scalogram is that of Table 22.1, but with each observation marked as to whether it conforms with its expectation or not. Observations within 0.5 rating points of their expectation are deemed to be in their expected categories, and are reported with their category values, e.g., '1', '2', etc. These ratings support the overall inferential relationship between observations and measures. Observations more than 0.5 rating points away from their expectations, i.e., in a "wrong" category, are marked with a letter equivalent: 'A' = '0','B' = '1','C' = '2', etc. These contradict observation-to-measure inferences. The proportion of in- and out-of-category observations are reported by the COHERENCE statistics in Table 3.2. |
Table 22.3 Guttman scalogram of original response codes
GUTTMAN SCALOGRAM OF ORIGINAL RESPONSES: KID |ACT |1111112 1 221 121 22 |8920311251427634569784035 |------------------------- 2 +2222222222222222222222222 M ROSSNER, LAWRENCE F. 41 +2222222222222222222222212 F FATALE, NATASHA 34 +2222222222222222222222211 F PASTER, RUTH 17 +2222222222222222222222210 F SCHATTNER, GAIL 50 +2222222222222222222122111 M DOEPPNER, TOM 45 +2222222222222222222221200 F MCLOUGHLIN, BILLY 7 +2222222222222222220220220 M WRIGHT, BENJAMIN 16 +2222222222222222222111101 F BUFF, MARGE BABY 48 +2222222222222222212221100 M CHAZELLE, BERNIE 25 +2222222222222222212021101 F SEILER, KAREN 59 +2222222222222222212111020 F CLAPP, LB 18 +2222222222222222212102100 M ERNST, RICHARD MAX |
The observations are printed in order of person and item measures, with most able persons listed first, the easiest items printed on the left. This scalogram shows the original codes in the data file. |
Interpreting a Scalogram. Example of Table 22.1 for Example0.txt. A scalogram orders the persons from high measure to low measure as rows, and the items from low measure (easy) to high measure (hard) as columns. The green boxes show the probabilistic advance from responses in high categories on the easy items to responses in low categories on the hard items.
Top left corner: where the “more able” (more liking) children respond to the “easier” (to like) items. So we expect to see responses of “Like” (2). We do! Unexpected responses in this area influence the Outfit statistics more strongly.
Top right corner: (blue box) where the “most liking” children and the “hardest to like items” meet - you can see some ratings of 1.
Bottom right corner: where “less able” (less liking) children respond to the “harder” (to like) items. So we expect to see responses of “Dislike” (0). But do we?? Something has gone wrong! There are 1’s and 2’s where we expected all 0’s. Unexpected responses in this area influence the Outfit statistics more strongly.
Diagonal transition zone: Between the red diagonals lies the transition zone where we expect 1’s. In this zone Infit is more sensitive to unexpected patterns of responses. More categories in the rating scale means a wider transition zone. Then the transition zone can be wider than the observed responses.
Example of Table 22.2. Five categories: 0,1,2,3,4. Categories in the wrong scores bands are shown as @ (=0), A (=1), B (=2), C (=3), D (=4). The 5 score bands have different colors here. Band 0 is yellow. Band 1 is green. Band 2 is blue. Band 3 is white. Band 4 is mauve.
Example of Table 22.3. Here is the Scalogram for Example 5, a computer-adaptive, multiple-choice test. The original responses are shown.
GUTTMAN SCALOGRAM OF ORIGINAL RESPONSES:
STUDENT |TOPIC
| 11 3 11 1 3212232212132425654421434145625 36555366356465464633654
|640215038189677748390992315641640517264579268221076889430372139553458
|---------------------------------------------------------------------
3 + d b c db d c b cd S STA
6 + c b c d d c cd S CHU
27 + c c c c cbb a dd c b dd S ERI
4 + c b a d cb c dd d b d dc S CHE
9 + d c a a cbc c ba b dca c db dc bbdbb c S MEL
25 + c b d c bc a a d a c dba d cbccbd c S DAN
24 + a d ac b c d c b a a AP SAR
12 + c b c c b b b b a cb cc c A ALL
28 + c c a bcc b c acc d b bc b c bdcb d IH GRE
21 + b c b a c b a bc bb ab a db bbd cc b b AP MIL
11 + a d c b cc cbad d ca c d ca ba b ac bac bc d c cc S CHR
17 + d d c c acac cac b c a a b c c bc b ac ac IH FRA
7 + d c d ccaaac d bcd ca b cb c b a a a a ab b bb ac IH JAN
1 + c c ab ca dc b a a c a c IM CAT
20 + bd b c b cad cd a bac b a b c b IL JAM
22 + d a d c b d cdabcc bbc cbdb a a dbac c dc d a IM BAR
30 + d a a cab cdd d d da c d c c db IL HOL
14 + b d c db caa caa da c dcd b c a cd d c c d b IL TAK
23 + d cab dcaa cadbbbba a c a db c b a c c ab c IL MEI
26 + d a cc b ac cdd b a d db a d b d d cb c IL MIT
15 + d d a a caba dcbddc c d b b a c bb NH MAR
13 + d d d c bac dc bd d dcd d a d d c b b NH KEL
29 +b dbd bdcabd baaacbd c d c ab a dd d c b c NH RIC
8 +bbdadbd abaa acbcbccacbcadab c d cdab a d a a c c cb IL ROB
16 +b da b bdccbbbcccbadaabba ccd c ac bd a a b NM HEL
18 +b a a d bca bc a cccc a db c a a d d NH JOE
10 +b badbdbcccadacbababcda cba a a acd a d a b NM JON
19 +bbdad dacbdb cabdbc c ad d ab b c a b NM THE
2 +bbda a d aaaa c c a cbd ab a d c NM KAT
5 + aad a ab a b a NL SMI
|---------------------------------------------------------------------
| 11 3 11 1 3212232212132425654421434145625 36555366356465464633654
|640215038189677748390992315641640517264579268221076889430372139553458
Example: To display Scalogram response strings on the person Tables 6, 17, 18,19.
1. Perform your standard Winsteps analysis.
2. Output Table 22, with a LINELENGTH= big enough for all observations to be on one line.
3. Save Table 22 to your Desktop (or wherever).
4. Open the Table 22 file in software that can do a rectangular copy (NotePad++, TextPad, Word, etc.)
5. Open your Winsteps data file using the same rectangular-copy software.
6. Rectangular copy the Scalogram immediately adjacent to the person labels.
7. Adjust Winsteps control NAMELENGTH= etc for the new data file format.
8. Save the control and data file(s)
9. Perform your revised Winsteps analysis.
10. Table 6 should now display the Scalogram as part of the person label.
Help for Winsteps Rasch Measurement and Rasch Analysis Software: www.winsteps.com. Author: John Michael Linacre
Facets Rasch measurement software.
Buy for $149. & site licenses.
Freeware student/evaluation Minifac download Winsteps Rasch measurement software. Buy for $149. & site licenses. Freeware student/evaluation Ministep download |
---|
Forum: | Rasch Measurement Forum to discuss any Rasch-related topic |
---|
Questions, Suggestions? Want to update Winsteps or Facets? Please email Mike Linacre, author of Winsteps mike@winsteps.com |
---|
State-of-the-art : single-user and site licenses : free student/evaluation versions : download immediately : instructional PDFs : user forum : assistance by email : bugs fixed fast : free update eligibility : backwards compatible : money back if not satisfied Rasch, Winsteps, Facets online Tutorials |
---|
Coming Rasch-related Events | |
---|---|
Jan. 17 - Feb. 21, 2025, Fri.-Fri. | On-line workshop: Rasch Measurement - Core Topics (E. Smith, Winsteps), www.statistics.com |
Feb. - June, 2025 | On-line course: Introduction to Classical Test and Rasch Measurement Theories (D. Andrich, I. Marais, RUMM2030), University of Western Australia |
Feb. - June, 2025 | On-line course: Advanced Course in Rasch Measurement Theory (D. Andrich, I. Marais, RUMM2030), University of Western Australia |
Apr. 21 - 22, 2025, Mon.-Tue. | International Objective Measurement Workshop (IOMW) - Boulder, CO, www.iomw.net |
May 16 - June 20, 2025, Fri.-Fri. | On-line workshop: Rasch Measurement - Core Topics (E. Smith, Winsteps), www.statistics.com |
June 20 - July 18, 2025, Fri.-Fri. | On-line workshop: Rasch Measurement - Further Topics (E. Smith, Facets), www.statistics.com |
Oct. 3 - Nov. 7, 2025, Fri.-Fri. | On-line workshop: Rasch Measurement - Core Topics (E. Smith, Winsteps), www.statistics.com |
Our current URL is www.winsteps.com
Winsteps® is a registered trademark