读写结合
40S.C.Weigle/AssessingWriting9(2004)27–55
Table2
SummaryofresultsfromGSTEPRegentsFall2001–Summer2003
SemesterAdministered
formNRhetoricLanguageSALPass(%)aFail-R(%)a
13.86
13.40
15.63
15.47
15.02
16.56
15.91
14.78
15.10
16.18
15.163.399(0.50)3.5170(0.69)4.0047(0.90)3.8865(0.90)3.8261(0.94)3.8343(0.96)3.8651(0.88)3.8321(0.91)3.8436(0.90)4.3228(1.00)4.0627(0.84)5(0.28)26(0.25)4(0.08)5(0.07)2(0.03)2(0.04)3(0.05)1(0.04)4(0.10)–2(0.06)Fail-LorB(%)a4(0.22)6(0.06)1(0.02)2(0.03)2(0.03)–4(0.07)1(0.04)––3(0.10)Fall2001Spring20021Globalization1813.672Computers10213.711Computers2Globalization5215.887215.716515.494516.535815.522315.744014.862816.253215.09Summer20021ComputersFall2002Spring20031Globalization2Globalization1Globalization2PopulationSummer20031Population2Population
a1.00=100%.
time.Thiswouldsuggestthatthelatterinterpretationismoreplausible:thatweaker
studentsmaypostponethetestwhenpossible.Inanycasethedatadonotshow
anyconsistentpatternofbiasduetotesttopic;thereisasmuchvariationwithin
topicsasacrosstopics.
ThetablealsoshowsthatscoresfromFall2001werenoticeablylowerthan
thoseinsubsequentsemesters.Onepossiblereasonforthisresultisthatthetest
wasnewandstudentsandteacherswereunfamiliarwiththetestformat.After
the rstsemester,morematerialsweredisseminatedtostudentsandinstructors
describedindetailthetestformatandthescoringrubric;thismayaccountforthe
increaseinoverallscoresfollowingthe rstsemester.
Intermsofreasonsforstudents’failing,thetableshowsthatstudentsweremore
likelytofailonthebasisofrhetoricthanonthebasisoflanguage.Themajorityof
studentswhofailedthetestdidsoonthebasisofrhetoric,particularlyinthe rst
yearofthetest.Languagewasanadditionalfactorinfailuresinsomecases,but
forthemostpartlanguageinandofitselfwasnottheprimaryreasonforfailure.
OnlyfourstudentspassedontheRhetoricscalebutfailedontheLanguagescale;
becauseofthesmallnumberofthesecasestheyarenotshowninthetable.
Inadditiontoconsistencyofscoringacrossprompts,anotherimportantaspect
ofreliabilityisinter-raterreliability,orthedegreetowhichratersagreeontheir
scores.Thisisfrequentlyreportedasareliabilitycoef cient;however,inpractical
terms,thepercentageofagreementbetweentworatersissometimesamoreuseful
indicatorofreliability.FortheAlternateRegentsTest,threeratersreadallessays
andscoredthemeither1(fail)or2(pass).Table3showsthepercentageofessays
onwhichthe rsttworatersagreedandthepercentageonwhichallthreeraters
agreed,bytestform.Asthetableshows,theoverallagreementratebetweenthe
rstandsecondraterswas79%,withratesvaryingbyformfrom58%to88%.The
此外,听力材料和阅读材料的输入可以:1)减少单模态(monomodal)输入 不足导致的写成绩欠佳的可能性;2)补充背景知识缺乏导致的测试真实性和公平性偏颇 (Weigle 2004)...
分项式评分标准更有利(如Bacha,2001;Hamp—Lyons,1991;Knoch,2009;Weigle,2002)...■● 16●’’● 2004)。 _ IIlIll● l—IIIII● I l——I●● IIIII●...
(2000) Assessing Writing Sara Cushing Weigle (2002) Assessing Listening Gary Buck (2001) Assessing Grammar James E Purpura (2004) Statistical Analyses for ...
Nunes (2004) carried out a project over the period of a year with a ...might restrain the performance of EFL writers (Hartman, 2001; Weigle, 2002...
事实上,大 学里的写作任务是基于阅读进行的,学生需要 借助于来自阅读材料的外界信息进行写作、尤 其是学术写作(Weigle,2004)。其次,语言测 北京第二外国语学院学报...
但是, 工作记忆对 书面语产出影响的研究相对较少 ( Olive,2004 ),代 表性 ...( Weigle,2005 ) 其二, 研究二语时, 工 作记忆 容量测量却使用母语, 如 ...
95 Feb. 2004 电子档案袋 ) ) ) 外语写作测评的新理念和新方法张红霞, 王...和档案袋写作测评( portfolio as sessment) 三个 模式( Weigle 2002: 239) 。...
J Gastrointest Surg 2004; 8: 398-400. 5. Cummings DE, Weigle DS, Frayo RS et al. Plasma ghrelin levels after diet-induced weight loss or gastric ...
Weigle B,Heizrnann CW.BiochemK B卸hys Commun.2004; 316(3):949-59.t4:ShishiboriT,OyamaY,MatsushitaO,YamashitaKobayashiR.BiochemZ1999;338rPt3):583.9....
(2004). Issues in validity in progressive paradigms of qualitative research....Weigle, S. C. (2010). Validation of automated scores of TOEFL iBT tasks...
热门文档