UML Di\DJUDPODUÕ.XOODQÕPÕQÕQ <D]ÕOÕP*HUHNOHUL
*|]GHQ*HoLUPH3HUIRUPDQVÕQD(WNLOHUL %LU
Replikasyon dDOÕúPDVÕ
Özlem Albayrak %LOJLVD\DU7HNQRORMLVLYH%LOLúLP6LVWHPOHUL øKVDQ'R÷UDPDFÕBilkent Üniversitesi Ankara, Türkiye [email protected]Özet. %XoDOÕúPDda \D]ÕOÕPJHUHNOHULVSHVLILNDV\RQXGRNPDQÕQDGDKLOHGLOHQ 80/GL\DJUDPODUÕQÕQ\D]ÕOÕPJHUHNOHULJ|]GHQJHoLUPHVUHFLQGHUDSRUODQDQ, YH GR÷UXRODUDN EHOLUOHQHQ KDWDVD\ÕODUÕQD HWNLOHULQL LQFHleyen kontrollü deney VRQXoODUÕ sunulPDNWDGÕU Bir replikasyon çaOÕúPDVÕ RODQ EX oDOÕúPDQÕQ DPDFÕ NRQX\OD LOJLOL GDKD |QFH \DSÕOPÕú özgün oDOÕúPD VRQXoODUÕQÕQ JHoHUOLOL÷LQL DUWÕUPDNWÕU. 5HSOLNDV\RQGD|]JQoDOÕúPDGDNXOODQÕODQODUGDQIDUNOÕPDWHU\DOOHU kulODQÕOPÕúWÕUdDOÕúPD\DNDWÕODQ66 VRQVÕQÕI OLVDQV|÷UHQFLVLQLQWDPDPÕ uygu-ODPDOÕ \D]ÕOÕP PKHQGLVOL÷L GHUVL YH EX GHUVLQ |QNRúXOX RODQ \D]ÕOÕP PKHQGLVOL÷LGHUVOHULQL]RUXQOXRODUDNDOPÕúWÕU
dDOÕúPD EXOJXODUÕna göre \D]ÕOÕP JHUHNOHUL belirtim GRNPDQÕQGa UML NXOODQÕP GXUXPODUÕ YH 80/ VÕQÕI GL\DJUDPODUÕQÕQ EXOXQPDVÕ NDWÕOÕPFÕODUÕQ \D]ÕOÕPJHUHNOHULJ|]GHQJHoLUPHVUHFLQGHUDSRUODGÕNODUÕKDWDVD\ÕVÕQÕDQODPOÕ olarak etkilemektedir. <D]ÕOÕP JHUHNOHULQL J|]GHQ JHoLUHQOHU 80/ GL\DJUDP ODUÕQ GDKLO HGLOPHVL GXUXPXQGD GDKD ID]OD VD\ÕGD GR÷UX KDWD EHOLUOHPHNWHGLU 80/ GL\DJUDPODUÕQÕQ dahil edilmesiyle elde edilen olumlu etki, akademik EDúDUÕQRWODUÕGúNRODQ|÷UHQFLOHUGHGDKDID]ODRODUDNJ|]OHPOHQPLúWLU'HQ H\GHNXOODQÕODQJHUHNVLQLPGRNPDQODUÕQÕQKDQJLVÕUDLOHNXOODQÕOGÕ÷ÕQÕQUDSRU ODQDQKDWDVD\ÕVÕVWQGHDQODPOÕELU HWNLVLYDUGÕU80/GL\DPODUÕQNXOODQÕOPDVÕ LOH KDWD EXOPDN LoLQ KDUFDQDQ ]DPDQ DUDVÕQGD da DQODPOÕ ELU LOLúNL gözlemlenPLúWLU +HP |]JQ KHP GH UHSOLNDV\RQ oDOÕúPDVÕ 80/ GL\DJUDP ODUÕQÕQ dahil eGLOPHVL GXUXPXQGD GDKD ID]OD VD\ÕGD KDWD UDSRUODQGÕ÷ÕQÕ göstermektedir.
Anahtar Kelimeler: \D]ÕOÕPJ|]GHQJHoLUPH; UML; gereksinim gözden geçirme
1
*LULú
IEEE Standard for Reviews and Audits 1028:2008’e göre gözden geçirme “bir \D]ÕOÕPUQQQKDWDEXOPDYH \D]ÕOÕPODLOJLOLVWDQGDUWYHVSHVLILNDV\RQODUGDQVDS PD GD GDKLO ROPDN ]HUH DQRUPDOOLNOHUL EHOLUOHPH DPDFÕ\OD J|UVHO RODUDN LQFHOHQ mesidir [1]. <D]ÕOÕP JHUHNOHUL ile ilgili hataODUÕQ belirlenmesi \D]ÕOÕP PKHQGLV OL÷LQGH\D]ÕOÕPNDOLWHJYHQFHVLWHNQLNOHUi içinde en etkin yöntemlerdendir [2, 3].
8QLILHG 0RGHOLQJ /DQJXDJH 80/ GL\DJUDPODUÕ \D]ÕOÕP JHOLúWLUPHGH VÕNOÕNOD NXOODQÕOÕU [5, 6@YHJ|UVHOOLNYHPRGHOOHPHGHVWH÷LYHULU>@1HVQHPRGHOOHULDQDOLVWOHU YH NXOODQÕFÕODU DUDVÕQGDNL LOHWLúLP SUREOHPOHULQi en aza indirger [5]. Pratikte, ELOLúLP sistemleri JHUHNVLQLPGRNPDQODUÕQGD80/GL\DJUDPODUÕQDUDVWODPDN\D\JÕQGÕU[7].
*HUHNVLQLPOHUGHNL KDWDODU LOHUOH\HQ DúDPDODUÕ GD ROXPVX] HWNLOH\HFH÷LQGen mali-\HWL \NVHN KDWDODUGÕU > @ Bu hatalar ]DPDQÕQGD belirlenmezse, ileride çok daha \NVHN PDOL\HWOHUH QHGHQ RODFDNODUGÕU <D]ÕOÕP J|]GHQ JHoLUPH DODQÕQGD \DSÕODQ ELOLPVHOoDOÕúPDODUÕQ ELULQFLODPDFÕ süreçlerin HWNHQOLNYHHWNLQOLNOHULQLQDUWÕUÕOPDVÕGÕU [9]. UML ve gözden geçirme ile ilgili olarak daha önce yapÕODQ DUDúWÕUPDODU 80/ GL\DJUDPODUÕ EXOXQDQ GRNPDQODUÕQ J|]GHQ JHoLULOPHVL VUHFLQLQ HWNLQOHúWLULOPHVLQH G|QN \ROODUÕQ EXOXQPDVÕQGD RGDNODQPÕúWÕU /LWHUDWUGH J|]GHQ JHoLUPH VUHFLQGH IDUNOÕ RNXPD WHNQLNOHULQLQ KDQJLOHULQLQ GDKD HWNLQ ROGX÷XQX LQFHOH\HQ oRN VD\ÕGD DUDúWÕUPD\DSÕOPÕúWÕU>@ $\QÕRNXPDWHNQL÷LNXOODQÕOGÕ÷ÕQGDELOHELUH\ VHOHWNLQOLNOHUDUDVÕQGDIDUNOÕOÕNODUJ|]OHPOHQPLúWLU>@
%X oDOÕúPDQÕQ DVÕO RGD÷Õ J|]GHQ JHoLUPH VUHFLQGH NXOODQÕODQ GRNPDQODUGÕU dDOÕúPDGDRNXPDWHNQL÷L YHJ|]GHQJHoLUHQOHULVDELWWXWXOPXú80/NXOODQÕPGXUX PX YHQHVQHGL\DJUDPODUÕLoHUHQYH\DLoHUPH\HQEHQ]HUbüyüklük ve NDUPDúÕNOÕNWD GRNPDQODU NXOODQÕOPÕúWÕU 80/ GL\DJUDPODUÕQ DQDOLVW YH NXOODQÕFÕODU DUDVÕQGDNL iletLúLP SUREOHPOHULQL D]DOWPDVÕQD benzer úHNLOGH >@, analist ve gözden geçirenler DUDVÕQGDNLLOHWLúLP SUREOHPOHULQLGHD]DOWDFD÷ÕQÕGúQ\RUX].
ÇDOÕúPDGD JHUHNVLQLP GRNPDQÕQD HNOHQHQ 80/ GL\DJUDPODUÕQÕQ gereksinim J|]GHQJHoLUPHVUHFLQHRNX\XFXRODUDNNDWÕODQELUH\OHULQHWNHQOLN YHHWNLQOLklerine RODQ HWNLVLQL LQFHOH\HQ NRQWUROO GHQH\OHU \DSÕOPÕúWÕU 5HSOLNDV\RQODU ELUH\VHO GHQ H\OHULQEXOJXODUÕQÕQJHQHOOHúWLULOPHVLQGHNXOODQÕODQ\DUDUOÕoDOÕúPDODUGÕU> @ YD]ÕOÕP PKHQGLVOL÷L DODQÕQÕQ ROJXQODúPDVÕ LoLQ GDKD ID]OD \DNÕQ UHSOLNDV\RQ oDOÕúPDVÕQÕQ \DSÕOPDVÕQÕ gerekir [17, 18]. dDOÕúPD |]JQ oDOÕúPD\D EHQ]HU \DNÕQOÕNWDGÕU Özgün oDOÕúPDGD ELOGLULOHQ ED]Õ geçerlilik tehditlerini D]DOWPD\Õ DPDoODPDNWDGÕU > @ <DSÕVDO ve içsel geçerlilik problemlerini azaltmak için NDUPDúÕNOÕN ve büyüklük DoÕVÕQGDQ ELUELULQH EHQ]HU |]JQ oDOÕúPDGD NXOODQÕODQGDQ IDUNOÕLNL\HQLGRNPDQKD]ÕUODQPÕúWÕU>@*|]GHQJHoLUPHVUHFLQGHRNX\XFXURO QH VDKLS NLúLOHULQ ELUH\VHO SHUIRUPDQVODUÕQÕ EHOLUOHPHGH NXOODQÕODQ GLO VHYL\HOHUL GH önemlidir [23]. Bu nedenlHUHSOLNDV\RQoDOÕúPDVÕQGD\HQLGRNPDQODUNDWÕOÕPFÕODUÕQ DQDGLOLRODQ7UNoH¶GHKD]ÕUODQPÕúWÕU
dDOÕúPD sunumunda deneysel replikasyon rehberi L]OHQPLúWLU >@ 2. bölümde JHoPLúWHNRQXLOHLOJLOL\DSÕODQ EHQ]HUoDOÕúPDODU|]HWOHQPLú 3. bölümde replikasyon WDVDUÕPÕQÕQ D\UÕQWÕODUÕ VXQXOPXúWXU. Veri analizi E|OPQQ DUGÕQGDQ . bölüm geçerlilik teditlerini, ve 6. bölüm de sonuçlar ve geleFHNoDOÕúPDODUÕQÕ DQODWPDNWDGÕU.
2
øOJLOLdDOÕúPDODU
80/ GL\DJUDPODUÕQÕQ gözden geçirme sürecine etkilerini DUDúWÕUDQ oDOÕúPD VD\ÕVÕ \HWHUOLGH÷LOGLU. Güncel ELUOLWHUDWUWDUDPDVÕQGDELOHEHQ]HUoDOÕúPDyoktur [24].
%LUH\VHO J|]GHQ JHoLUPH VUHFLQLQ NDOLWHVLQL DUWÕUPDN LoLQ DUDúWÕUPDFÕODU oRNOXNOD RNXPDWHNQLNOHULVWQGHRGDNODQPÕúODUGÕU>@+DQJLRNXPDWHNQL÷LQLQ GDKD L\L ROGX÷X LOH LOJLOL NHVLQ ELU RUWDN VRQXo HOGH HGLOPHPLú ROPDVÕQD NDUúÕQ NXOODQÕPGXUXPODUÕLoHUHQGRNPDQODUÕQHWNLQOL÷LQLDUWÕUPDNDPDFÕ\OD\HQLELURNXPD WHNQL÷LRODUDN.XOODQÕP7DEDQOÕ2NXPD 8VDJH%DVHG5HDGing-8%5 JHOLúWLULOPLúWLU [8]. UBR önceliklendLULOPLú NXOODQÕP GXUXPODUÕQÕ HVDV DOÕU %X oDOÕúPDGD NXOODQÕP GXUXPODUÕ GL\DJUDPODUÕ DoÕNODPD LoHUPHGL÷LQGHQ 8%5 L]OHQPHPLúWLU <D]ÕOÕP J|]GHQJHoLUPHLOJLOL\DSÕODQoDOÕúPDODUÕQ|QHPOLELUNÕVPÕJ|]GHQJHoLUPHVUHFLLOH ilgilidir. UM/GL\DJUDPODUÕQÕQ J|]GHQJHoLUPHLOHLOJLOLELUGH÷LúNHQRODUDNLQFHOHQ PHVL NRQXVX OLWHUDWUGH HOH DOÕQPDPÕúWÕU > @ Bununla birlikte, \D]ÕOÕPEDNÕPDúDPDVÕLOHLOJLOL\DSÕODQoDOÕúPDODUGHQH\VHORODUDNJUDILNHOHPDQODUÕQ ID\GDODUÕQÕDUDúWÕUPÕúODUGÕU> @dDOÕúPDODUÕQ VRQXoODUÕQDJ|UHJUDILNHOHPDQODUÕQ YDUOÕ÷Õ PLPDULQLQGDKDL\LDQODúÕOPDVÕQÕYH EDNÕPoDOÕúPDODUÕQÕolumlu etkiler [28]. 80/GL\DJUDPODUÕQÕQ \DUDUODUÕLOHLOJLOLELOJLOHUHXODúÕODELOLU>-30, 36].
80/ GL\DJUDPODUÕQÕQ JHUHNVLQLP GRNPDQODUÕQGD \HUDOPDVÕQÕQ UDSRUODQDQ KDWD VD\ÕVÕQÕ DQODPOÕ HWNLOHGL÷L ELOGLULOPLúWLU >@ .DWÕOÕPFÕODU 80/ GL\DJUDPODUÕQÕQ YDUOÕ÷ÕQGDGDKDID]ODVD\ÕGDKDWDUDSRUODPÕúODUGÕU80/GL\DJUDPODUÕQÕQ GR÷UXRODUDN EHOLUOHQHQKDWDVD\ÕVÕQDDQODPOÕELUHWNLVLEXOXQPDPÕúWÕU>@
Özgün Deney
g]JQoDOÕúPDGDDNDGHPLNRUWDPGDG|UGQFVÕQÕIlisans |÷UHQFLVLQLQNDWÕOGÕ÷Õ deney \DSÕOPÕúWÕU [21]+HU|÷UHQFLLNLJHUHNVLQLPGRNPDQÕQÕLQFHOHPLúYHEXOGXNODUÕ KDWDODUÕ UDSRUODPÕúGÕU dDOÕúPDGD “bireysel gereksinim gözden geçirme sürecinde 80/ GL\DJUDPODUÕQÕQ JHUHNVLQLP GRNPDQÕQGD \HU DOPDVÕ”: H1:“raporlanan hata VD\ÕVÕQÕ´ + ³EXOXQDQ KDWD VD\ÕVÕQÕ´ + ³\DQOÕú SR]LWLI KDWD VD\ÕVÕQÕ´ YH + ³KDWD EXOPDNLoLQJHUHNHQ]DPDQÕ´HWNLOHUKLSRWH]OHULLQFHOHQPLúWir.
%D÷ÕPVÕ] GH÷LúNHQ: Gereksinim GRNPDQÕQÕQ 80/ GL\DJram içerip içermemesi, bD÷ÕPOÕGH÷LúNHQOHU: UDSRUODQDQKDWDVD\ÕVÕ, YHGR÷UXRODUDNbelLUOHQHQKDWDVD\ÕVÕGÕU.
3
5HSOLNDV\RQ7DVDUÕPÕ
$UDúWÕUPD6RUXODUÕ
S1) %LUH\VHO KDWD UDSRUODPD RUDQÕ, gereksinim doküPDQÕQGD 80/ GL\DJUDP NXOODQÕOPDVÕQGDQHWNLOHQLUPL"
S2) %LUH\VHO KDWD EHOLUOHPH RUDQÕ, JHUHNVLQLP GRNPDQÕQGD 80/ GL\DJUDP NXOODQÕOPDVÕQGDQHWNLOHQLUPL"
S3) %LUH\VHORODUDNGR÷UXúHNLOGHEHOLUOHQHQ|QHPVHYL\HVL\NVHNKDWDVD\ÕVÕ, ger-HNVLQLPGRNPDQÕQGD 80/GL\DJUDPNXOODQÕOPDVÕQGDQLOHHWNLOHQLUPL"
S4) Bireysel olarak kaydedilen \DQOÕúSR]LWLIVD\ÕVÕ, JHUHNVLQLPGRNPDQÕQGD80/ GL\DJUDPNXOODQÕOPDVÕQGDQHWNLOHQLUPL"
S5) *HUHNVLQLP KDWDVÕQÕQ EXOXQPD ]DPDQÕ JHUHNVLQLP GRNPDQÕQGD 80/ GL \DJUDPNXOODQÕOPDVÕQGDQHWNLOHQLUPL"
'H÷LúNHQOHU
%D÷ÕPVÕ] GH÷LúNHQ, HasUML, JHUHNVLQLP GRNPDQÕQÕQ 80/ GL\DJUDP LoHUGL÷L GXUXPGDLoHUPHGL÷LGXUXPGDGH÷HULQLDOÕU %D÷ÕPOÕGH÷LúNHQWDQÕPODUÕ:
5DSRUODQDQKDWDVD\ÕVÕ (RD): +HUELUNDWÕOÕPFÕWDUDIÕQGDQUDSRUODQDQWRSODP hata VD\ÕVÕ
'R÷UX RODUDN EHOLUOHQHQ KDWD VD\ÕVÕ (CDD): %LUH\VHO J|]GHQ JHoLUHQ WDUDIÕQGDQ GR÷UXRODUDNEXOXQDQWRSODPKDWDVD\ÕVÕHWNHQOLN|OoV
gQHPOL KDWD VD\ÕVÕ (HSD): %LUH\VHO J|]GHQ JHoLUHQ WDUDIÕQGDQ GR÷UX RODUDN EXOXQDQGL÷HUKDWDWUROPD\DQWRSODPKDWDVD\ÕVÕHWNHQOLN|OoV
<DQOÕú SR]LWLI VD\ÕVÕ (FP): %LUH\VHO J|]GHQ JHoLUHQ WDUDIÕQGDQ ND\GHGLOHQ KDWDOÕ \DQÕWODU%XGH÷LúNHQLQ\NVHNGH÷HUDOPDVÕ istenmez.
+DWD EHOLUOHPH ]DPDQÕ (AT): 'RNPDQ EDúÕQD J|]GHQ JHoLUPH LoLQ KDUFDQDQ ]D man. Bu çalÕúPDGD]DPDQGH÷LúNHQLJ|]|QQHDOÕQPÕúWÕU [27].
Hipotezler
Yerden kazanmak için VÕIÕU KLSRWH]OHULDúD÷ÕGDki hipotezler listesine eklenmedi: H1: *HUHNVLQLP GRNPDQÕQGD bulunan 80/ GL\DJUDPODUÕQÕQ YDUOÕ÷Õ UDSRUODQDQ KDWDVD\ÕVÕQÕ (RD) etkiler.
H2: Gereksinim dokümDQÕQGD EXOXQDQ 80/ GL\DJUDPODUÕQÕQ YDUOÕ÷Õ GR÷UX EXOXQDQKDWDVD\ÕVÕQÕ (CDD) etkiler.
+ *HUHNVLQLP GRNPDQÕQGD EXOXQDQ 80/ GL\DJUDPODUÕQÕQ YDUOÕ÷Õ bulunan |QHPOLKDWDVD\ÕVÕQÕ (HSD) etkiler.
H4: *HUHNVLQLP GRNPDQÕQGD EXOXQDQ 80/ GL\DJUDPODUÕQ YDUOÕ÷Õ \DQOÕú SR]LWLI VD\ÕVÕQÕ (FP) etkiler.
H5: *HUHNVLQLP GRNPDQÕQGD EXOXQDQ 80/ GL\DJUDPODUÕQ YDUOÕ÷Õ doküman gözden geçirme süresini (AT) etkiler.
.DWÕOÕPFÕODU
.DWÕOÕPFÕODU|]HOELUQLYHUVLWHQLQOLVDQVSURJUDPÕQGD]RUXQOXELUGHUVRODQ8\JX ODPDOÕ<D]ÕOÕP0KHQGLVOL÷LGHUVLQHND\ÕWOÕ 66 DGHWG|UGQFVÕQÕI|÷UHQFLVLdir. Tümü J|QOO RODUDN NDWÕOPÕúWÕU .DWÕOÕPFÕODUÕQ Lú deneyimi ortalama D\GÕU *özden geçirme verilen H÷LWLPOH VD÷ODQPÕúWÕU g÷UHQFLOHU 80/ GL\DJUDPÕ oL]PH YH \D]ÕOÕP JHUHNVLQLPGRNPDQÕ KD]ÕUODPDNRQXVXQGDEDúODQJÕç seviyesinde bilgi sahibidir.
'HQH\+D]ÕUOÕ÷ÕYH8\JXODPDVÕ
Deney öncesinde |÷UHQFL\e JHUHNVLQLP GRNPDQÕ J|]GHQ geçirme süreci ile ilgili olarak GDNLNDVUHQELUH÷LWLPYHULOPLúWLU(÷LWLPGHRNXPDWHNQLNOHULKDNNÕQGD biOJL YHULOPLú YH GHQH\OH LOJLOL DoÕNODPDODU \DSÕOPÕúWÕU 'HQH\H NDWÕODFDN JUXSODUÕQ
ROXúWXUXOPDVÕQGD NDWÕOÕPFÕODUÕQ 80/ LOH LOJLOL |Q WHVWWH ve akademik performans GH÷HUOHULYHGHQH\LPOHULJ|]|QQHDOÕQPÕúWÕU. 'HQH\H|÷UHQFLNDWÕOPÕúWÕU
'HQH\ LNL DúDPDGD JHUoHNOHúWLULOPLúWLU +HU ELU DúDPD GDNLND LOH VÕQÕU ODQGÕUÕOPÕúWÕU. dDOÕúPDQÕQ WRSODP süresi, H÷LWLPL GH LoHUHFHN úHNLOGH, 110 dakika ROPXúWXU 'HQH\LQ WDVDUÕPÕ 7DEOR ’de gösterilmektedir. +HU DúDPD VRQUDVÕQGD PD WHU\DOOHUWRSODQPÕú\HQLOHULGD÷ÕWÕOPÕúWÕU
Tablo 1. 'HQH\$úDPDODUÕ
(÷LWLPYHgQWHVW GDNLND
$úDPD, GDNLND Diyagram Tipi
UML No-UML
Sistem A 3$56ø0B: QFD Grup 1Grup 4 Grup 2Grup 3
$úDPD,, GDNLND Diyagram Tipi
UML No-UML
Sistem A 3$56ø0B: QFD Grup 3Grup 2 Grup 4Grup 1
.XOODQÕODQ0DWHU\DO
dDOÕúPDQÕQ|QWHVWNÕVPÕQGD-25*,6YHH÷LWLPNÕVPÕQGD &DXVH(IIHFW*UDphing 7RRO &(*7 LVLPOL IDUNOÕ gereksinim dokümanlarÕ NXOODQÕOPÕúWÕU 'HQH\ LoLQ JQFHO OHQPLú B: Quality Function Deployment (QFD) ve A: Particle Simulator Software (ParSIM) dokümanlaUÕNXOODQÕOPÕúWÕU.DWÕOÕPFÕODUGRNPDQODUÕQLOJLOLROGX÷X konulara WDQÕGÕN GH÷LOOHUGL+HUELUVLVWHPLoLQ80/GL\DJUDPODUÕLoHUHQYHLoHUPH\HQVUPOHUL ROPDN ]HUH LNLúHU JHUHNVLQLP GRNPDQÕ ROXúWXUXOPXú, 80/ GL\DJUDPODUÕQÕn hiçbiri LoLQ D\UÕQWÕOÕ DoÕNODPD YHULOPHPLú \DOQÕ]FD GL\DJUDPODU HNOHQPLúWLU .XOODQÕODQ VLV WHPOHUNDUPDúÕNOÕNYHER\XWDoÕVÕQGDQEHQ]HUGLU ve ilgili standarda [35] uyumludur.
Deneyde kullanÕODQPDWHU\DOOHUsiVWHP\D]ÕOÕPJHUHNVLniPGRNPDQÕ, dokümanda EXOXQDQKDWDODUÕQEHOLUOHQPHVLQGHNXOODQÕODFDNNRQWUROOLVWHVLYHKDWDODUÕQND\GHGLOPH VLQGH NXOODQÕODFDN KDWa ND\ÕW IRUPXQGDQROXúPDNWDGÕU .DWÕOÕPFÕODUÕQGHQH\ VRQUDVÕQ GDNL J|]OHP YH \RUXPODUÕQÕ DOPDN LoLQ GHQH\ VRQUDVÕQGD ELU anket \DSÕOPÕúWÕU DokümDQODUD VHUSLúWLULOHQ KDWD WUOHUL HNVLNOLN EHOLUVL]OLN WXWDUVÕ] ELOJL KDWDOÕ ELOJL JHUHNVL]ELOJLYHGL÷HUROPDN]HUHDOWÕVÕQÕIWDQROXúPDNWDGÕU>@
7DVDUÕP
'HQH\JHUoHNOHúWLULOPHGHQ|QFHNDWÕOÕPFÕODUD80/ELOJLOHULQL|OoHQELU|Q-test uy-JXODQPÕúYHJ|]GHQJHoLUPHVUHFLLOHLOJLOLRODUDNGDNLNDH÷LWLPYHULOPLúWLU+HUELU NDWÕOÕPFÕQÕQ LNL D\UÕ Vistem gHUHNVLQLP GRNPDQÕQÕ J|]GHQ JHoLUPHVLQL PPNQ NÕOPDN DPDFÕ\OD D\UÕ JUXS ROXúWXUXOPXúWXU Sistemlerin UML içeren ve içermeyen YHUVL\RQODUÕ LOH Woplam 4 DGHW JHUHNVLQLP GRNPDQÕ 7DEOR ’deki gibi DWDQPÕúWÕU %|\OHFH KHU ELU NDWÕOÕPFÕ hem UML içeren ve LoHUPH\HQ LNL D\UÕ GRNPDQÕ J|]GHQ JHoLUHELOPLúWLU. 7DVDUÕP ROXúDELOHFHN|÷UHQPHHWNLVLni GHD]DOWPÕúWÕU
4
Veri Analizi ve Sonuçlar
'HQH\ JUXSODUÕQGD \HUDODQ |÷UHQFLOHULQ GD÷ÕOÕPÕ QRW RUWDODPDODUÕ YH |Q WHVWWHQ DOGÕNODUÕVRQXoODUDJ|UHEDúODQJÕoWDKRPRMHQRODUDNGD÷ÕWÕOPÕúODUGÕ1RWRUWDODPDODUÕQÕ (CGPA) JHQHODNDGHPLNEDúDUÕVÕYH|QWHVWVRQXoODUÕQÕ80/ELOJLOHULQL|OoPHDPDFÕ\ ODNXOODQGÕN%XGH÷HUOHULNXOODQDUDNJUXSROXúWXUXOGX$QFDNH÷LWLPHNDWÕODQGRNX] |÷UHQFLQLQGHQH\HNDWÕOPDPDVÕQHGHQL\OHJUXSODUDUDVÕQGDSODQODQDQGHQJHGHQVDSPD ROXúWX Tablo 2 5'GH÷LúNHQLiçin betimsel istatistik GH÷HUOHULQLJ|VWHUPHNWHGLUDen-H\HNDWÕODQNDWÕOÕPFÕ$129$YDUVD\ÕPODUÕQÕVD÷ODGÕ> @.
Tablo 25'gUQHNOHP7DQÕPÕ1
Min Max Averaj Sapma Min Max Averaj Sapma
RD 2,00 27 10,03 4,894 HSD- 0 10 2,69 2,281
CDD 0 17 6,74 3,778 FP- 0 20 3,29 3,529
.XOODQÕODQ GRNPDQ 80/ GL\DJUDP LoHUGL÷LQGH haFPL DUWPDVÕQD NDUúÕQ NDWÕOÕPFÕODU GRNPDQÕQ 80/ LoHUPHGL÷L GXUXPD RUDQOD GDKD ID]OD KDWD rapor-ODPÕúODUGÕU ùHNLO 6DGHFH 5' GH÷LO D\QÕ ]DPDQGD &'' GH÷LúNHQL LoLQ GH 80/ GL\DJUDPÕ LoHUHQ GRNPDQODU LQFHOHQGL÷LQGH GDKD ID]OD VD\ÕGD GR÷UX KDWD WHVSLW ediOPLúWLU ùHNLO 2). 7HVSLW HGLOHQ WP KDWDODUÕQ |QHP GHUHFHVL D\QÕ GH÷LOGLU >@ dDOÕúPDGD ³dL÷HU” hata tipi DOWÕQGD, genellikle \D]ÕP LOH LOJLOL KDWDODUÕn, öteki hata tiplerine nazaran “daha az önemli” ROGXNODUÕ YDUVD\ÕOGÕ. Bu nedenle HSD GH÷HUL KHVDSODQÕrken, “GL÷HU” türüne giren hatalar NDSVDPGÕúÕQGDWXWulPXúWXU
Not ortalaPDODUÕ ED]OÕ ROXúWXUXODQ KHU ELU JUXS LoLQGH 80/ GL\DJUDP EXOXQDQ GRNPDQGD GDKD ID]OD KDWD ROGX÷XQX UDSRUODPÕúWÕU ùHNLO 80/ GL\DJUDPODUÕQÕQ UDSRUODQDQ KDWD VD\ÕVÕQD RODQ NDWNÕVÕ \NVHN DNDGHPLN RUWDODPD GH÷HULQH VDKLS olan |÷UHQFLOHU DUDVÕQGD HQ D]GÕU. øNLQFL J|]GHQ JHoLUPHGH GDKD D] VD\ÕGD hata rapor-ODQPÕúWÕU ùHNLO .
dDOÕúPDGD \HUDODn tüm istatistik testlerde alfa düzeyi NXOODQÕOGÕ .DWÕOÕPFÕODU DUDVÕQGDNL WHVWOHU GRNPDQODUÕQ LQFHOHQPH VÕUDVÕ YH DNDGHPLN RUWDODPD JUXEX GH÷LúNHQOHULQLQ 5' GH÷LúNHQL ]HULQGH DQODPOÕ HWNLVL ROGX÷XQX J|VWHUGL 7DEOR YH Tablo 4 $129$WHVWVRQXoODUÕQÕDQODWDQWDEORODUGDúXNÕVDOWPDODU NXOODQÕOPÕúWÕU KT: .DUHOHU7RSODPÕP Ortalama, S DQODPOÕOÕNGHUHFHVL, *$ *UXSODU$UDVÕ*ø *UXSøoL T: Toplam 0,00 0,25 0,50 0,75 1,00 HasUM L 5 10 15 20 25 RD A A A A A P Q 0,00 0,25 0,50 0,75 1,00 HasUM L A A
0,00 0,25 0,50 0,75 1,00 HasUML 0,00 10,00 20,00 CDD A A A A A A 0,00 0,25 0,50 0,75 1,00 HasUML 0,0 2,5 5,0 7,5 10,0 HS D A A A A A ùHNLO 2+DV80/%D]ÕQGD&DD ve HSD
$129$ WHVW VRQXoODUÕ GRNPDQ LQFHOHPH VÕUDVÕ )129,1=31,233, p=0,000) ve DNDGHPLN QRW RUWDODPDVÕ JUXEX )118,3 S GH÷LúNHQOHULQLQ 5' LOH DQ ODPOÕLOLúNLVLROGX÷XQXJ|VWHUPLúWLUDRNPDQLQFHOHPHVÕUDVÕYH akademik performans GH÷LúNHQOHULLOLúNLOLGH÷LOGLU
5' 5DSRUODQDQ KDWD VD\ÕVÕ A129$ WHVWLQLQ VRQXoODUÕ +DV80/ GH÷LúNHQLQLQ UDSRUODQDQ KDWD VD\ÕVÕ GH÷LúNHQL VWQGH DQODPOÕ ELU HWNLVL ROGX÷XQX J|VWHUPLúWLU ) S .DWÕOÕPFÕODUJ|]GHQJHoLULOHQGRNPDQ80/GL\DJUDPODU LoHUGL÷LQGHGDKDID]ODVD\ÕGDKDWDUDSRUODPÕúODUGÕr (Tablo 5).
&'' 'R÷UXEHOLUOHQHQKDWDVD\ÕVÕ): $129$WHVWLQLQVRQXoODU+DV80/GH÷LúNHni LOH &'' GH÷LúNHQL DUDVÕQGD DQODPOÕ ELU LOLúNL ROGX÷XQX ELOGLUPHNWeGLU .DWÕOÕPFÕODU LQFHOHQHQGRNPDQODUÕQ80/GL\DJUDPLoHUPHVLGXUXPXQGDGDKDID]ODVD\ÕGDGR÷UX hatD\ÕEHOLUOHPLúOHUGLU(F129,1=7,253, p=0,008) (Tablo 7).
+6' %XOXQDQ|QHPOLKDWDVD\ÕVÕ $129$WHVWLQLQVRQXoODUÕ+DV80/GH÷LúNHQ LQLQ +6' GH÷LúNHQL VWQGH DQODPOÕ ELU LOLúkisi olmadÕ÷ÕQÕ J|VWHUPHNWHGLU 7DEOR ). $QODPOÕ ROPa\DQ EX LOLúNLQLQ QHGHQL EHOLUOHQHQ D] |QHP GHUHFHVLQH VDKLS KDWDODUÕQ ND\GHGLOPHVLQHGHQL\OHGL÷HUKDWDODUÕQWHVSLWHGLOPHVLLoLQNDODQ]DPDQÕQVÕQÕUOÕROPDVÕ olabilir. .DWÕOÕPFÕODU 80/ GL\DJUDP LoHUHQ GRNPDQODU LQFHOHGLNOHULQGH 80/ LoHUPH\HQGRNPDQDQD]DUDQGDKDID]ODVD\ÕGD|QHPOLKDWDEHOLUOHPLúOHUGLU ùHNLO
)3 <DQOÕú3R]LWLI): $129$WHVWVRQXoODUÕ+DV80/GH÷LúNHQLQLQ)3]HULQGHDQ ODPOÕ ELU HWNLVL ROPDGÕ÷ÕQÕ |QHUPLúWLU )3 )129,1=2,250, p=0,136). (Tablo 9). .DWÕOÕPFÕODU80/GL\DJUDPODUGDKLOHGLOGL÷LQGHGDKDID]ODVD\ÕGD)3rDSRUODPÕúODUGÕU ùHNLO
AT(Harcanan Zaman): *|]GHQ JHoLUPH VUHFLQGH KDUFDQPDVÕ LoLQ EHOLUOHQHQ ]D PDQÕQ VW OLPLWL ROPDVÕQD NDUúÕQ $129$ WHVW VRQXoODUÕ 80/ GL\DJUDPODUÕQ GDKLO edilmesinin gözden geçirme sürecinde harcanan zaman üstünde pozitif ve anlamOÕELU HWNLVLROGX÷XQX|QHUPLúWLU(F126,1=25,564, 0,000) (Tablo 10). (ùHNLO
Tablo 3. RD ve CGPA-Group ANOVA
KT df m F p
GA 190,125 3 63,375 2,776 ,044 *ø 2625,152 115 22,827
T 2815,277 118
Tablo 4. RD ve CEGT-Total ANOVA
KT df m F p
GA 95,696 5 19,139 ,793 ,557 *ø 3018,182 125 24,145
Tablo 5. RD ve HasUML ANOVA KT df m F p GA 231,209 1 231,209 10,347 ,002 *ø 2882,669 129 22,346 T 3113,878 130 Tablo 65'YH6ÕUD$129$ KT df m F p GA 608,957 1 806,957 31,233 ,000 *ø 2506,921 129 19,433 T 3113,878 130 5 10 15 20 25 RD A A A 1,00 2,00 3,00 4,00 A A 0,00 0,25 0,50 0,75 1,00 HasUM L 5 10 15 20 25 RD A S S A A 0,00 0,25 0,50 0,75 1,00 HasUM L A A ùHNLO 3$NDGHPLN3HUIRUPDQV*UXSODUÕYH+DV80/'H÷LúNHQOHUL%D]ÕQGD5' 1 2 Order 5 10 15 20 25 RD A A 0 1 1 2 Order A A A
'HQH\VRQUDVÕ\DSÕODQDQNHWHJ|UHNXOODQÕODQPDWHU\DOQH]RUQHGHNROD\GÕU, süre X\JXQGXU .DWÕOÕPFÕODU 80/ GL\DJUDPODUÕQ J|]GHQ JHoLUPH VUHFLQGH ID\GDOÕ ROGXNODUÕQÕEHOLUWPHOHULQHNDUúÕQGL\DJUDPODUÕD\UÕQWÕOÕRODUDNLQFHOHPHPLúOHrdir. Tablo 7. CDD ANOVA KT df m F p GA 98,761 1 98,761 7,253 ,008 *ø 1756,415 129 13,616 T 1855,176 130 Tablo 8. HSD ANOVA KT df m F p GA 3,267 1 3,267 ,626 ,430 *ø 672,901 129 5,216 T 676,168 130 0,00 0,25 0,50 0,75 1,00 HasUML 0,0 2,5 5,0 7,5 10,0 HS D A A A A A ùHNLO+DV80/%D]ÕQGD+6' Tablo 9. FP ANOVA KT df m F p GA 27,749 1 27,749 2,250 ,136 *ø 1591,228 129 12,335 T 1618,977 130 0 1 HasUML 400 800 1200 1600 Ti m e A A ùHNLO. +DV80/%D]ÕQGD$7 0,00 0,25 0,50 0,75 1,00 HasUML 0,00 5,00 10,00 15,00 20,00 FP A A A A A ùHNLO+DV80/%D]ÕQGD)3 TABLO 10. AT ANOVA KT df m F p GA 4573363 1 4573363,200 25,564 ,000 *ø 22540886 126 178895,924 T 27114250 127
dDOÕúPD6RQXoODUÕQÕQ.DUúÕODúWÕUÕOPDVÕ
g]JQ oDOÕúPDGD ROGX÷X JLbi 80/ GL\DJUDPODUÕQÕQ JHUHNVLQLP GRNPDQÕQD HNOHQPHVLQLQUDSRUODQDQKDWDVD\ÕVÕVWQGHDQODPOÕ ve pozitif etkisi gözlemlenmiúWLU.
g]JQ oDOÕúPDGD +DV80/¶LQ &'' GH÷LúNHQL VWQGH DQODPOÕ ELU HWNLVL EXOXQ PDPÕúWÕUreplikasyonda ise +DV80/LOH&''DUDVÕQGDDQODPOÕLOLúNLJ|]OHPOHQPLúWLU
5HSOLNDV\RQ oDOÕúPDVÕQGD WDQÕPODQDQ +6' YH )3 GH÷LúNHQOHUL VWQGH +DV80/ GH÷LúNHQLQLQDQODPOÕHWNLVLJ|]OHPOHQPHPLúWLUgWH\DQGDQ+DV80/GH÷LúNHQLQLQ$7 GH÷LúNHQLVWQGH pozitif ve DQODPOÕHWNLVLYDUGÕU
Akademik not orWDODPDVÕED]ÕQGDWPJUXSODU80/GL\DJUDPÕLoHUHQGRNPDQODUÕ NXOODQGÕNODUÕQGD GDKD ID]OD VD\ÕGD KDWD UDSRUODPÕúODUGÕU 5DSRUODQDQ KDWD VD\ÕODUÕ DUDVÕQGDNL IDUN DNDGHPLN SHUIRUPDQVÕ HQ GúN RODQ JUXSWD HQ \NVHNWLU $NDGHPLN SHUIRUPDQVÕHQ\NVHNRODQJUXpta ise bu fark minimumdur.
5
Geçerlilik Tehditleri
øo JHoHUOLOLN ELU oDOÕúPDQÕQ VRQXoODUÕQÕQ WDVDUÕPÕQGDNL KDWDODUD GH÷LO GH PDQLSOH HGLOHQ ED÷ÕPVÕ] GH÷LúNHQOHUH ED÷ODQDELOPH GHUHFHVL GÕú JHoHUOLOLN JHQHOOHúWirme ile ilgilidir [34]. <D]ÕOÕPPKHQGLVOL÷LLOHLOJLOLoDOÕúPDODUGD|÷UHQFLOHULQNDWÕOÕPFÕROPD ODUÕ\D\JÕQGÕU>@ g÷UHQFLOHULQNDWÕOPDVÕQÕQFLGGLELUWHKGLWROPDGÕ÷ÕQÕGúQ\RUX] .DWÕOÕPFÕODU DODQ ELOJLVL \HULQH 80/ PRGHOOHPH ELOJLVLQH WDQÕGÕN ELUH\OHUGLU g]JQ GHQH\OH UHSOLNDV\RQ oDOÕúPDVÕ DUDVÕQGDNL HQ |QHPOL IDUNÕQ L\LOHúWLUPHQLQ NXOODQÕODQ PDWHU\DOOHU E\NON YH NDUPDúÕNOÕN NDWÕOÕPFÕODUÕQ DODQ ELOJLVLQH RODQ \DNÕQOÕNODUÕ YHGHQH\|QFHVLQGHYHULOHQH÷LWLPLOHGHQH\VRQUDVÕYH|QFHVL\DSÕODQWHVWOHUROGX÷XQX bildirmeliyiz. Replikasyon çDOÕúPDVÕQGDNL PDWHU\DOOHU |]JQ oDOÕúPDGDNLQGHQ IDUNOÕGÕU )DUNOÕPDWHU\DOOHULQNXOODQÕOPDVÕGÕúJHoHUOLOL÷LQLQDUWÕUÕOPDVÕQGD\DUGÕPFÕGÕU. 2NXPDWHNQL÷LJ|]GHQJHoLUPHVUHFLQLQHWNLQOL÷LQGH|QHPOLELUGH÷LúNHQGLU> 8, 10]. )DUNOÕELU RNXPDWHNQL÷LQLQNXOODQÕOPDVÕJHoHUOilLNWHKGLWOHULQLD]DOWDFDNWÕU
dDOÕúPDX\JXODPDVÕQGDGHQH\LWDVDUOD\DQYHVRQXoODUÕQÕGH÷HUOHQGLUHQ\D]DUGÕúÕQ GD oDOÕúPD\OD GR÷UXGDQ LOJLVL ROPD\DQ J|QOO LNL IDUNOÕ |÷UHWLP HOHPDQÕQÕQ görev DOPDVÕ deneyi yaSDQ NLúL |Q\DUJÕVÕQÕ D]DOWPÕúWÕU. Ö÷UHWLP HOHPDQODUÕ ve NDWÕOÕPFÕlar |÷UHQFLOHUoDOÕúPDQÕQKLSRWH]OHULYHDUDúWÕUPDVRUXODUÕKDNNÕQGDELOJLVDKLELGH÷LOGLUOHU <DSÕVDOJHoHUOLNLOHLOJLOLRODUDNLNLQFLDúDPDGD|÷UHQFLOHULQELUNÕVPÕQÕQGHQH\L erken WDPDPODPDH÷LOLPGHROPDVÕJ|]OHPOHQPLúWLU
6
6RQXoYH*HOHFHNdDOÕúPDODU
dDOÕúPDVRQXoODUÕ|]JQoDOÕúPDGDUDSRUODQDQKDWDVD\ÕVÕ GH÷LúNHQLLOHLlgili bulgu-ODUÕGHVWHNOHUNHQGR÷UXEHOLUOHQHQKDWDVD\ÕVÕ GH÷LúNHQLLOHLOJLOLEXOJXODUODoHOLúPHNWH dir. 'RNPDQ KDFPLQGH GL\DJUDPODUÕQ GDKLO HGLOPHVL\OH DUWÕúD NDUúÕQ D\QÕ ]DPDQGD EXOXQDQGR÷UXKDWDVD\ÕVÕGL\DJUDPODUÕQGDKLOHGLOGL÷LGXUXPODUGDGDKDID]ODGÕURep-OLNDV\RQGD \HQL GH÷LúNHQOHU KDWD WLSOHUL |QHP GHUHFHOHUL \DQOÕú KDWD VD\ÕVÕ YH NXOODQÕODQ ]DPDQ GD LQFHOHQPLúWLU dDOÕúPDGD J|]GHQ geçirenlerin etkinlikleri incele-QHQ GRNPDQ 80/ LoHUGL÷LQGH DQODPOÕ RODUDN DUWDUNHQ YHULPOLOLNOHUL DQODPOÕ RODUDN D]DOPÕúWÕU.DWÕOÕPFÕODU80/LoHUHQGRNPDQLQFHOHPHOHULQGHGDKDID]OD]DPDQKDU FDPÕúYHGDKDID]ODVD\ÕGDKDWDUDSRUODPÕúODUGÕU(WNLQOLNDUWÕPÕ80/¶OHULQJHUHNOHULQ DQODúÕOPDODUÕQD RODQ NDWNÕODUÕ LOH LOLúNLOHQGLULOHELOLUNHQ YHULPOLOLNWHNL D]DOPD GD GL
\DJUDPODUÕQ RNXQPDVÕ LoLQ KDUFDQDQ HN ]DPDQ NXOODQÕPÕ LOH DoÕNODQDELOLU Kesin ve JHQHOOHúWLUHELOLUsonuçlar elde edilmesi için daha fazla sayÕGDYHIDUNOÕortamlarda rep-OLNDV\RQ oDOÕúPDODUÕQÕQ \DSÕOPDVÕ JHUHNPHNWHGLU dDOÕúPDGD NXOODQÕODQ PDWHU\DOOHULQ TürNoHYHøQJLOL]FHYHUVL\RQODUÕLoLQ\D]DULOHED÷ODQWÕ\DJHoLOPHVLQL|QHUL\RUX]
7
7HúHNNU
Marcela Genero Bocco(UKDQ<FHHUYH7XQD.ÕOÕo¶D \RUXPODUÕ, Duygu Albayrak, Syed A. Ali YHNDWÕOÕPFÕODUDoDOÕúPDQÕQJHUoHNOHúPHVLQGHNLNDWNÕODUÕLoLQWHúHNNUOHU.
8
Kaynaklar
[1] S. Engineering, S. Committee, and I. Computer, IEEE Std 1028TM-2008 (Revision of IEEE Std 1028-1997), IEEE Standard for Software Reviews and Audits, vol. 2008, no. August. 2008.
[2] A. Aurum, H. Petersson, and C. Wohlin, 2002. “Stateof- the-art: Software Inspections After 25 Years,” Software Testing, Verification and Reliability.vol. 12, 2002, pp.133-154.
[3] A. Porter, L. Votta, V. Basili, “Comparing Detection Methods for Software Requirements Inspections: A Replicated Experiment,” IEEE Trans. on Software Engineering, vol. 21, 1995, pp.563-575.
[4] Booch, G., J. Rumbaugh, and I;. Jacobson, The Unified Modeling User Guide. Addison Wesley, 1999.
[5] A. Endres and D. Rombach, A Handbook of Software and System Engineering: Empirical Observations, Laws and Theories, IESE, 2003.
[6] S. N. Bhatti, “Why Quality? ISO 9126 Software Quality Metrics ( Functionality ) Support by UML Suite,” Software Engineering Notes, vol. 30, no. 2, pp. 1-5, 2005.
[7] G. Kösters, Hans-Werner S., and M. Winter, “Coupling Use Cases and Class Models as a Means for Validation and Verification of Requirements Specifications,” Requirements Engineering, vol. 6, no. 1, pp. 3-17, Feb. 2001.
[8] Thelin, T., Runeson, P. Wohlin, C., “Prioritized Use Cases as a Vehicle for Software Inspections”, IEEE Software, vol 20, no.4, 30-33, 2003.
[9] F. Salger, G. Engels, and A. Hofmann, “Inspection effectiveness for different quality attributes of software requirement specifications: An industrial case study,” 2009 ICSE Workshop on Software
Quality, pp. 15-21, May. 2009.
[10] J. C. Carver, N. Nagappan, and A. Page, “The Impact of Educational Background on the Effectiveness of Requirements Inspections: An Empirical Study,” vol. 34, no. 6, pp. 800-812, 2008. [11] R. Conradi, P. Mohagheghi, T. Arif, L. C. Hegde, G. A. Bunde, and A. Pedersen, “Object-Oriented
Reading Techniques for Inspection of UML Models – An Industrial Experiment,” ECCOP pp. 483-500, 2003.
[12] K. Cox and` R. Jeffery, “An Experiment in Inspecting the Quality of Use Case Descriptions,”
Practice, vol. 36, no. 4, pp. 211-229, 2004.
[13] Basili V., Shull F. and Lanubile F. (1999) Building knowledge through families of experiments, IEEE Transactions on Software Engineering, 25(4), 435-437, 1999.
[14] F. Shull, J. Carver, S. Vegas, N. Juristo. The Role of Replications in Empirical Software Engineering.
Empirical Software Engineering Journal. 13:211–218. 2008.
[15] Juristo, N. and Vegas, S. Using differences among replications of software engineering experiments to gain knowledge. In Anonymous ESEM '09: Proceedings of the 2009 3rd International Symposium
on Empirical Software Engineering and Measurement. (). IEEE Computer Society, Washington, DC,
USA, 356-366, 2009.
[16] B. Kitchenham. The role of replications in empirical software engineering—a word of warning.
[17] Vegas, S., Juristo, N., Moreno, A.M., Solari, M., and Letelier, P. 2006. Analysis of the Influence of Communication between Researchers on Experiment Replication. ISESE 2006. 28-37.
[18] F. Shull, J. Carver,G.H. Travassos, J.C. Maldonado, R. Conradi and V.R. Basili, Replication Studies: Building a Body of Knowledge About Software Reading Techniques, in Lecture Notes on Empirical Software Engineering eds. N. Juristo and A. M. Moreno, 39-84, 2003.
[19] Cruz-Lemus, J. a, Genero, M., Manso, M. E., Morasca, S., & Piattini, M. (2009). “Assessing the understandability of UML statechart diagrams with composite states—A family of empirical studies”.
Empirical Software Engineering, 14(6), 685-719. doi:10.1007/s10664-009-9106-z
[20] Carver, J. C. "Towards reporting guidelines for experimental replications: A proposal." In Proceedings of the 1st International Workshop on Replications in Empirical Software Engineering (Held during ICSE). May 4, 2010.
http://cs.ua.edu/~carver/Papers/Conference/2010/2010_RESER.pdf
[21] Ö. Albayrak, “An Experiment to Observe the Impact of UML Diagrams on the Effectiveness of Software Requirements Inspections,” ESEM 2009, pp. 506-510.
[22] M. Genero, and M. Piattini, ”Empirical validation of measures for class diagram structural complexity through controlled experiments”, 5th International ECOOP Workshop on Quantitative Approaches in Object-Oriented Software Engineering, 2001.
[23] Carver, J. C. (2003) PhD. Thesis. The Impact of Background and Experience on Software Inspections. University of Maryland, 2003.
[24] Mohagheghi, P., Dehlen, V., & Neple, T. (2009). Definitions and approaches to model quality in model-based software development – A review of literature. Information and Software Technology,
51(12), 1646-1669. Elsevier B.V. doi:10.1016/j.infsof.2009.04.004
[25] Laitenberger, O., Atkison, C., El-Emam, K. “Using Inspection Technology in Object-oriented Development Projects,” NRC/ERB-1077, no. June, 2000.
[26] Bratthall, L., & Wohlin, C. (2002). Is it possible to decorate graphical software design and architecture models with qualitative Information?-An experiment. IEEE Transactions on Software Engineering, 28(12), 1181-1193. doi:10.1109/TSE.2002.1158290
[27] Ricca, F., Penta, M. D., Torchiano, M., Society, I. C., Tonella, P., & Ceccato, M. (2010). How Developers’ Experience and Ability Influence Web Application Comprehension Tasks Supported by UML Stereotypes: A Series of Four Experiments. Computer, 36(1), 96-118.
[28] M. Staron, L. Kuzniarz, and C. Thurn. An Empirical Assessment of Using Stereotypes to Improve Reading Techniques in Software Inspections. in 3-WoSQ: Third Workshop on Software Quality.2005. St. Louis, USA: ACM.
[29] Arisholm, E., Briand, L. C., Member, S., Hove, S. E., & Labiche, Y. (2006). The Impact of UML Documentation on Software Maintenance: An Experimental Evaluation, IEEE Trans. On Software
Engineering 32(6), 365-381.
[30] Dzidek, W. J., Arisholm, E., & Briand, L. C. (2008). A realistic empirical evaluation of the costs and benefits of UML in software maintenance. IEEE Transactions on Software Engineering, 34(3), 407-432.
[31] Carver, J., Jaccheri, L., Morasca, S., Shull, F., Software, E., Group, E., & Science, I. (n.d.). Using Empirical Studies during Software Courses, 81-103. 2004 ESERNET 2001-2003LNCS 2765 [32] F.J. Gravetter and L.B. Wallnau, Statistics for the Behavioural Sciences, 7thedition, 2007.
[33] D.E. Harter, C.F. Kemerer, and S.A. Slaughter, “Does Software Process Improvement Reduce The 6HYHULW\RI'HIHFWV"$ /RQJLWXGÕQDO)LHOG6WXG\´,(((7UDQV2Q6RIWZDUH(QJLQHHULQJDFFHSWHG paper.
[34] Wohlin, C., P. Runeson, M. Höst, M. C. Ohlsson, B. Regnell, and A. Wesslen, Experimentation in
Software Engineering: An Introduction, Kluwer Academic Publishers, 1999.
[35] IEEE Recommended Practice for Software Requirements Specifications, IEEE Std 830-1998, 25 August 1998.
[36] .ÕOÕo g 6D\ B., Demirörs, O. Cognitive Aspects of Error Finding on a Simulation Conceptual Modeling Notation, Computer and Information Sciences, ISCIS 2008, pp.1-6.
