TRNG I HC Y DC TP H CH MINH KHOA Y T CNG CNG B mn Thng k Y Hc v Tin Hc STATA NG DNG TRONGNGHIN CU KHOA HC Bin son: TS. Vn Dng THNH PH H CH MINH 4-2008 (Lu hnh ni b) MC LC i cng v thng k v thng k m t ...................................................................................... 1 Mt s nh ngha ....................................................................................................................... 1 Bin s v cc loi bin s.......................................................................................................... 1 Phng php m t tm tt v trnh by s liu ......................................................................... 1 Cc s thng k m t................................................................................................................. 2 Phng php trnh by s liu .................................................................................................... 5 i cng v phn tch s liu...................................................................................................... 13 Suy lun thng k...................................................................................................................... 14 Ci t chng trnh Stata 10.0, s liu muv cc chng trnh c lin quan.......................... 29 Khi ng v kt thc Stata.......................................................................................................... 36 Khi ng Stata ............................................................................................................................ 40 1. Khi ng Stata .................................................................................................................... 40 2. M t giao din ca chng trnh Stata ................................................................................ 40 3. Cch cch thc hin lnh trong chng trnh Stata ......................................................... 41 4.Lu li kt qu phn tch ..................................................................................................... 42 Mt vi phn tch n gin vi Stata............................................................................................ 44 M t s liu vi Stata 10.0 for Windows .................................................................................... 58 Thng k phn tch bin s nh lng vi Stata ......................................................................... 96 Thc hnh ................................................................................................................................. 98 1 i cng v thng k v thng k m t Mt s nh ngha Thng k l phng php khoa hc dng thu thp, tm tt, trnh by v phn tch s liu. S liu: Kt qu c c do vic quan st hay thu thp mt bin s cc i tng khc nhau hay thi gian khc nhau. Th d:Khi ti quan st gii tnh ca cc hc vin trong lp, ti c s liu l: Nam, nam, n, n, n, nam, n, v.v Th d: Mt nh nghin cu o nng hemoglobin ca 70 thai ph c kt qu nh sau: 10.213.710.414.911.512.011.0 13.312.912.19.413.210.811.7 10.610.513.711.814.110.313.6 12.112.911.412.710.611.411.9 9.313.514.611.211.710.910.4 12.012.911.18.810.211.612.5 13.412.110.911.314.710.813.3 11.911.412.513.011.613.19.7 11.215.110.712.913.412.311.0 14.611.113.510.913.111.812.2 v nhng con s ny c gi l s liu. Cn lu s liu phi lin kt vi mt bin s nht nh. Nu ti quan st gii tnh ngi ny, tui ca ngi khc, qun o ca mt ngi khc na th kt qu quan st c khng phi l s liu. Bin s v cc loi bin s Bin s l nhng i lng hay nhng c tnh c th thay i t ngi ny sang ngi khc hay t thi im ny sang thi im khc. Nh vy bin s c th th hin i lng hay c tnh.-Nubinsthhinmtilngncgilbinsnhlng(quantitative variable). Bin s nh lng c th cn c chia thnh bin s t s - ratio variable(c gi tr khng tuyt i) v bin s khong interval variable (khng c gi tr khng tuyt ) - Nu bin s nhm th hin mt c tnh, bin s c gi l bin s nh tnh. Bin s nh tnh cn c chia lm 3 loi:- Bin s nh gi binary variable (khi ch c 2 gi tr)- Bin s danh nh nominal variable (khi c 3 hay nhiu hn cc gi tr v cc bn thn cc gi tr khng c tnh cht th t) - Bin s th t - ordinal variable (khi c 3 haynhiu hn cc gi tr v cc bn thn cc gi tr c tnh cht th t- Ngoi ra c khi bin c khng ch c quan tm v phng din n c xy ra hay cha xy ra m cn c quan tm v phng din bin c xy ra vo lc no. Th d sau khi iu tr bnh nhn ung th chng ta khng ch quan tm bnh nhn c t vong hay khng m cn quan tm bnh nhn bnh nhn t vong bao nhiu lu sau khi iu tr v nu bnh nhn cha t vong, bnh nhn sng c bao lu.Phng php m t tm tt v trnh by s liu 2 -T l cho tng gi tr nul bin th t hay danhnh- T l cho gi tr tiu biunu l bin nh gi-T l cho tng gi tr nul bin th t hay danhnh- T l cho gi tr tiu biunu l bin nh gi Cc s thng k m t C hai loi thng k m t: thng k m t khuynh hng tp trung v thng k m t tnh phn tn. Thng k m t khuynh hng tp trung Thng k m t khuynh hng tp trung c th l trung bnh (mean), trung v (median) v yu v (mode). Nhng thng k ny cho bit gi tr tiu biu cho s liu. Th d: c hai loi thuc h p A v B. Gi s c 5 i tng sau khi s dng thuc h p A s c huyt p 110 - 115 -120 - 125 -130 v 5 i tng khc sau khi s dng thuc h p B s c huyt p 120 - 125 - 130 - 135 - 140. Con s tiu biu nht cho bit tc dng cathuc A l huyt p trung bnh sau khi s dngthucAvl120.Conshuytptrungbnhnythphnhuytptrungbnhsaukhisdng thuc B cho bit thuc A c tc dng mnh hn. Trung bnh ca s liu, c k hiu l (x (c l x gch) l tng cc gi tr ca s liu chia cho s ln quan st (N). Nxxi=Th d: S liu v huyt p tm thu ca 5 i tng l 120, 125, 130, 135, 150. Huyt p tm thu trung bnh s l 132 1325150 125 130 125 120=+ + + +==Nxxi 3 Dokhngththchinccphptonshctrnccbinsnhtnh(danhnhvtht) chng ta ch c th tnh trung bnh cho s liu ca bin s nh lng. Nu chng ta sp xp s liu theo th t, gi tr ng gia c gi l trung v. Nu c hai gi tr cng ng gia, trung bnh cng ca hai gi tr ny l trung v. Th d: S liu v huyt p tm thu (mmHg) ca 5 i tng l 120, 125, 130, 135, 150. Trung v ca huyt p tm thu l gi tr ng gia v bng 130 S liu v chiu cao (cm) ca 6 ngi l 153, 155, 160, 162, 165, 161. tnh trung v, trc tin chng ta phi sp xp s liu ny: 153, 155, 160, 161, 162, 165. Do c hai gi tr 160 v 161 cng gia, trung v s l (160+161)/2 = 160,5 cm Do bn cht cabin s danh nh khng th sp c theo th t, chng ta ch c th tnh trung v ca s liu nh lng v s liu ca bin s th t. Ngoi ra yu v (mode) cng c s dng lm con s thng k tiu biu. Yu v l gi tr xut hin ph bin nht (c tn sut cao nht). Thd:Sliuvhuytptmthu(mmHg)ca5itngl120,125,130,135,150.Trong trng hp ny khng c yu v. im s ca 5 hc sinh l 5, 5, 6, 7, 9. Yu v ca im s l 5. Trong mt p c 361 gia nh ngi Kinh, 120 gia nh ngi Khmer v 27 gia nh ngi Hoa. Yu v ca bin s dn tc l dn tc Kinh. Trong mt s liu c th, c th khng c yu v, c th c mt yu v hoc hai hay nhiu yu v. y l khuyt im chnh ca s thng k ny. Do vy ngi ta thng ch dng yu v cho bin s danh nh hay trong cc trng hp c bit C th s dng trung bnh, trung v hay yu v cho bin s nh lng. Khi bin s nh lng c phnphibnhthng(hnhchung)thbaconsnyxpxbngnhauvkhingita thng tnh trung bnh bi v trung bnh c nhng c tnh ton hc mnh. Tuy nhin nu s liu b lch th con s trung v phn nh gi tr tiu biu mt cch chnh xc hn. Thd:Bnhnhnblotddy-ttrngciutrtheomtphcditvikhun Helicobacter. Sau iu tr, bnh nhn c theo di v ghi nhn thi gian k t khi s dng thuc n lc bt u ci thin triu chng au. 10 bnh nhn thi gian ny (ngy ) l nh sau: 1, 2, 2, 2, 2, 2, 3, 3, 3, 30. Bnh nhn c thi gian t lc iu tr n lc gim triu chng l 30 ngy trn thcchtlbnhnhnkhngpngviiutr.Trungvvtrungbnhcasliul2v5 ngy. Con s trung v phn nh chn thc hn bi v vi t cch l mt bc s lm sng t s liu trn c th nhn xt rng mt bnh nhn tiu biu s gim au sau 2 ngy dng thuc. Con ss 30 trong th d trn c gi l s ngoi lai (outlier) v lm s liu b lch. Nhn chung, khi s liu b lch th con s trung bnh s b nh hng rt nhiu v khng phn nh gi tr tiu biu nh con s trung v. Thng k m t tnh phn tn: C3thngkmttnhphntn:lchchun,khongtphnvvphmvicasliu. Vic la chn thng k m t tnh phn tn c trnh by trong bng 2. Thngkmttnhphntnctmquantrngthhaisauconsmtkhuynhhngtp trung.Th d: Thuc h p A c s dng trn 5 bnh nhn v huyt p tm thusau khi dng thuc l 110, 115, 120, 125 v 130. Thuc h p B c s dng trn 5 bnh nhn v c huyt p sau s dng thuc l 100, 110, 120, 130, 140.Nh vy hai thuc h p ny c hiu qu h p l tng ng (bi v trung bnh ca hai s liu l bng nhau) nhng kt qu ca thuc B phn tn hn v iu ny lm thuc B tr nn km an ton. lch chun (standard deviation - vit tt l SD hay s) l con s nh gi mc phn tn v 4 c tnh theo cng thc: Nh vy lch chun phn nh khong cch trung bnh ca s liu so vi gi tr tiu biu. Khi nim lch chun ch c th p dng cho bin s nh lng bi v chng ta c th thc hin cc php ton s hc trn cc i lng nhng khng th thc hin trnccgi trca bin s nh tnh l cc c tnh. Thd:Sliuvhuytptmthu(mmHg)ca5itngl120,125,130, 135,150. Trungbnhca huyt p l 132 v lch chun bng 5 , 11 5 , 13245304324 9 4 49 1441 5) 132 150 ( ) 132 135 ( ) 130 132 ( ) 132 125 ( ) 132 120 (1) (2 2 2 2 212= = =+ + + += + + + + == =niiNx xs Phngsaivmttnguynlbnhphngcalchchun.Phngsai(variance)cth c k hiu v Var hay s2 v c tnh theo cng thc sau: ==niiNx xs1221) ( Phm vi ca s liu l tt c cc gi tr ca s liu t gi tr nh nht n gi tr ln nht. Th d: S liu v huyt p tm thu (mmHg) ca 5 i tng l 120, 125, 130, 135, 150. Phm vi ca bin s huyt p l 120 n 150. Th d: Thuc h p A c s dng trn 5 bnh nhn v huyt p tm thusau khi dng thuc l 110, 115, 120, 125 v 130. Thuc h p B c s dng trn 5 bnh nhn v c huyt p sau s dng thuc l 100, 110, 120, 130, 140. S liu ca thuc B ctnh phn tn cao hn do phm vi thay i t 100-140 trong khi phm v ca s liu thuc A ch t 110-130. Khongtphnv(inter-quartile):Nuchngtachiasliusptheothtlm2phnu nhau, khong t phn v l khong cch ca trung v phn trn v trung v phn di. Th d: S liu v huyt p tm thu (mmHg) ca 5 i tng l 120, 125, 130, 135, 150. S liu ny c chia lm 2 phn: phn 1 gm 120, 125, 130 v phn 2 gm 130, 135, v 150. Trung v ca phn trn l 125 - trung v ca phn di l 135, do phm t phn v l 125-135. Do bn cht ca khong t phn v l trung v ca phn s liu trn v phn s liu di, cng gingnhtrungv,khongtphnvkhngbnhhngbiccgitrngoilainhtrong trng hp ca lch chun. Cng nh trung v, khong t phn v ch c th p dng cho bin s nh lng hay th t. Cu hi: Phn tch trn my tnh v bin s hemoglobin cho kt qu sau. Hy th c v l gii kt qu: Variable | ObsMean Std. Dev. MinMax -----------+----------------------------------------------------- hemoglobin |7011.98429 1.4161228.8 15.1 ==niiNx xs121) ( 5 Phng php trnh by s liu S liu c th c trnh by thnh bng hoc cc th. Trnh by bng: Phn phi tn sut ca bin s nh tnh S liu ca bin s ri rc c th c trnh by di dng mt phn phi tn sut. Phn phi tn sut l mt bng ch ra tn sut xut hin ca tng gi tr ri rc ca bin s (Bng 1). Nh vy bng phn phi tn sut gm 2 ct, mt ct lit k cc gi tr ca bin s v mt ct trnh by tn sut tng ng ca cc gi tr . Table 1. Phn phi gii tnh ca 69 hc sinh lp cm thng trng mm non 23 thng 11, Huyn Hc mn GiiS trPhn trm Nam4565% N2435% Tng s69100% Bng trn l bn phn phi tn sut ca gii tnh. Bi v gii tnh c 2 gi tr nam v n nn ta lit k 2 gi tr ny mt ct. ct th nh ta ghi tn sut tng ng ca cc gi tr ny. i khi bng phn phi tn sut c thm ct phn trm nh trong th d trn. Bng 2 l mt th d khc v bng phn phi tn sut. Table 2. Phng php ca 600 tr trong bnh vin Phngphp S sinhPhn trm Sinh thng47879,7 Sinh forceps6510,8 Sinh m579,5 Tng s600100,0 Phn phi tn sut ca bin s nh lng Nubinslbinslintcchngtakhngthlitkttcccgitrcabins.Trong trng hp ny chng ta c th nhm (lm trn) gi tr ca bin s li. C th cc bc xy dng bng phn phi tn sut cho bin s nh lng nh sau: 1- Tm phm vi (gi tr cc tiu v gi tr cc i) ca s liu. Trong th d v hemoglobin ca 70 ph n phm vi l 8,8 n 15,1 2. Chia phm vi s liu ra lm n khong vi rng ca mi khong l d. Cn lu rng mi khong d nn l i lng chn nh 1, 2, 5, 10 hay 0,5, 0,2 v s cc khong n nn t 5-12 (trung 6 bnh l 7-8). Trong th d trn ta c th chia phm vi ra lm 8khong vi chiu rng khong bng 1nv.Khicckhongl:8-8,9;9-9,9;10-10,9;11-11,9;12-12,9;13-13,9;14-14,9;15-15,9. 3. m cc gi tr thch hp vo khong nh trc Hemoglobin (g/100ml) m 8-8,91 9-9,9111 10-10,9111111111111 11-11,91111111111111111 12-12,9111111111111 13-13,911111111111 14-14,91111 15-15,91 4.Xydngbngphnphitnsutvibinsvcckhonggitrcabinsvtnsut tng ng vi cc khong gi tr . Chng ta cng c th thm vo ct phn trm v ct phn trm tch ly(nu thch hp) Table 3. Hemoglobin ca 70 ph n HemoglobinTn sutPhn trmPhn trm tch ly 8-8,911.431.43 9-9,934.295.71 10-10,91420.0025.71 11-11,91927.1452.86 12-12,91420.0072.86 13-13,91318.5791.43 14-14,957.1498.57 15-15,911.43100.00 Th d nh nu bin s l chu vi vng cnh tay ca tr chng ta c th lm trn chu vi vng cnh tay n 1 cm. Khi ta c th xem thang o ca bin s l ri rc v trnh by bng phn phi tn sut ca bin s (bng 2). Table 4. Phn phi s o vng cnh tay ca 69 tr lp cm thng nh tr 23 thng 11, Hc mn. Vng cnh tayTn sutPhn trmPhn trm tch ly 7 13-
