This section is organized into a discussion of general issues and a list of specific data quality checks.

Size: px

Start display at page:

Download "This section is organized into a discussion of general issues and a list of specific data quality checks."

Jeremy Doyle
5 years ago
Views:

1 PANCEA Maual Qualit Assurace Checks This sectio is orgaized ito a discussio of geeral issues ad a list of specific data qualit checks. Geeral Issues The followig represet several importat geeral checks o the data. Complete the istrumets We eed a respose to each for ever questio asked, eve if this respose is Not Applicable (N/A) or "Do't Kow" (DK). Blak cells should impl that there is a pla to ask these questios later, ot that the have simpl bee forgotte. For a fuller discussio of data codes, see the maual chapter Data Collectio Completeess with actual data is most essetial for forest issues such as data o ke outputs. DK for tree issues (e.g., cost details) are udesirable, but havig some is acceptable. Provide adequate detail Eough iformatio should be provided to accuratel determie costs ad outputs for subsequet data cleaig ad aalsis. For example: Capital goods ad recurret goods represetig $1000 or more i spedig should be sufficietl detailed to later determie market prices from alterative suppliers. See Costig Techiques for more discussio of market pricig. Recurret services that are cotracted out should be explaied well eough to idicate the ature of the services provided. E.g., Provided a workshop for VCT couselors icludig refresher traiig for all couselors ad a advaced traiig for supervisors would be sufficietl detailed, whereas Traiig workshop would ot. If the resource cotet of a cotract is available, please iclude this as well (e.g., 70% persoel wages; 20% per diems ad 10% supplies). Persoel job categories ad titles should have explaator otes as eeded to esure that we have a good idea of the resposibilities that each perso carries. Medical Director at a STI cliic would be sufficietl iformative whereas MD would ot as good. Esure agreemet across sectios ad istrumets Data should be logicall cosistet withi the same istrumet ad across istrumets. For example: The work start date give o the Persoel Roster of the ARQi caot be later tha the work ed date. ARQi ad ADCi cost data should be idetical if the were tabulated for the same period of time ad for the same item. ADC itervetio outputs should be cosistet with the output data provided i HIPPI II ad i ARQ itervetio sheets. G-1 Versio 2/6/2004 PAN_ma_QA_v4

2 PANCEA Maual ARQi should have costs for supplies, such as codoms or test kits, the use of which is documeted i ARQ itervetio sheets. Base data are o writte records as much as possible Writte records should be the primar source of iformatio whe available. Data source iformatio should be complete so that there is iformatio o how data were collected ad a idicatio of its qualit (e.g., i geeral, writte records are more accurate tha recall). See the Data Collectio chapter for detail o data sources. Provide uits If uits of a item are ambiguous, tpes ad quatities should be specified. For example, whe prices are i a currec other tha the local currec, the currec tpe should be oted i the commets fields. Aother example: if 500 boxes of codoms were doated, we eed to kow how ma idividual codoms are i each box. Specific qualit checks I this sectio, we list specific data qualit assurace checks that the data teams should do withi ad betwee istrumets. We have doe this for ARQi, ARQsti, ad ARQvct. The qualit checks for other ARQ itervetio sheets ad for ADC will be similar i stle, ad largel overlappig due to shared questios. Items i bold are of particular importace. These are forest issues ad are usuall critical to the cost-effectiveess ratio or to a basic uderstadig of the itervetio s services ad their effectiveess. See especiall Q10 i ARQsti ad Q13 ad 1Q4 i ARQvct. These questios pertai to the are the ke outputs for these two itervetios. See 1 ote Q # Qualit Assurace Check? ARQ i 0.Cover A respose is give for each questio, eve if the respose is "do't kow" or "ot applicable". 1.Facilit Characteristics 4. The date is more recet tha the date give i Q3. The selected itervetios iclude those idicated i HIPPI, part 1, Q1 8.? & Q 9. The date is the same or more recet tha the date give i Q4. 1 "DB Validatio" meas that the MS Access database will be able to do the qualit assurace automaticall. "" = es. "" = o. "b" = both data collectors ad the MS Access database should examie. G-2 Versio 2/6/2004 PAN_ma_QA_v4

3 PANCEA Maual 10. The selected itervetios iclude those checked i HIPPI, part 1, Q1? & Q 11. The date give is i the past (or earlier this ear). 15. The aswer is < = 16. The aswer is < = 24. Make sure the aswer applies to the i 20. The area should be the catchmet area from which most patiets come; ot the etire coutr. 21. The umber is < = to the umber give i Q20, for each colum. 2 The umber is < = to the umber give i Q19, for each colum. A respose is give for each questio, eve if the respose is "do't kow" or "ot applicable". Goverace 5. If 6. ever, the Q6 should be 0. b 9. If the aswer to Q8 is "es", a appropriate descriptio of performace measuremet is give here. A respose is give for each questio, eve if the respose is "do't kow" or "ot applicable". 3a Buildigs 5. This area must be smaller tha the figure give i Q6. 8. The date give is equal to or more recet tha the date give i "Facilit Characteristics" Q3. 10, 15 Ret or mortgage should match the values i questio 5 of the ADCi, Buildig sheet. 13. The date give is equal to or prior to the date give i Q The date give is equal to or prior to the date give i Q The aswer give to Q29 is "es", if the aswer give to Q25 & Q26 is "es" for a buildig. b 30.. A source is provided if the aswer to Q29 is "es". 31. The aswer is < = The aswer is < = A respose is give for each questio, eve if the respose is "do't kow" or "ot applicable" , 2 ad 3. 3b.Persoel Roster Differet staff with the same iitials are marked with a "(1)", "(2)", etc. after the relevat iitials. The same staff member with multiple rows of data is give the same iitials i all rows of data. A clear descriptio is coveed of the overall resposibilities of each staff member. Am acroms that we are likel to be ufamiliar with are explaied (e.g., TLS=TB Liaiso Supervisor ad DDHS=Director of District Health Services ) voluteers ad traiees have a job categor describig their resposibilities accompaig their title. G-3 Versio 2/6/2004 PAN_ma_QA_v4

4 PANCEA Maual ad 3. staff members whose job titles have chaged over time have multiple rows of data. The date give is equal to or more recet tha the date give i "Facilit Characteristics" Q4. 9. If a amout is provided i Q10, the aswer here is "es". 10. The amout give is i local currec. If ot, a ote should documet the tpe of currec used People with higher job classificatios should almost alwas be receivig higher pa tha those with a lower classificatio. Where this is ot true, a brief ote is eeded. If the aswer to Q15 is "es", the the aswer give here is "es" as well. 1 The aswer is > 0 ad < = 60. If a aswer outside this rage is give, a explaatio is provided. 1 The aswer is > 0 ad < = 60. If a aswer outside this rage is give, a explaatio is provided. 21. % allocatios sum to 100%, for each row of data. 2 If es for a row, Q23 ad Q24 should be NA 18., 19. Efforts were made to complete these questios as much as possible. Not all aswers are "do't kow". If most aswers are "do't kow", a explaatio is provided regardig wh these data were ot collected. A respose is give for each questio, eve if the respose is "do't kow" or "ot applicable". 3c. Persoel turover The date give is toda's date or i the past., 6. The date give i Q2 + the time give i Q6 approximates "3c Persoel Roster", Q7 (the start date) for the replacemet staff member. 5. If o for a row, the Q6 is NA for that row. The aswer give is roughl o the same scale as "3b(1) Staff Roster" Q10 for the replacemet staff member. ocatios are a ver importat item as these figures have a 21. large effect o how itervetio costs are ultimatel tabulated. Please sped sufficiet time to get ver good estimates of these data. A respose is give for each questio, eve if the respose is "do't kow" or "ot applicable". 3d.Capital Goods Descriptios are detailed eough to cost the item at a store later. 1. Detailed descriptios are ol eeded o items o which $1,000 or more were spet i 12 moths. 1. Data were collected from facilit records as much as possible. 1. No items have bee double-couted. The date give is equal to or more recet tha the date give i "Facilit Characteristics" Q3. 5, 8, 11, 12 The amout give is i local currec. G-4 Versio 2/6/2004 PAN_ma_QA_v4

5 PANCEA Maual If ot, a ote should documet the tpe of currec used. 5. Value of each piece of equipmet is the value of the item whe doated/purchased as opposed to a curret valuatio 8. The aswer is "/a" if the aswer to Q7 is 0%. 10. If the aswer to Q9 is "es", tha the proportio is > 0%. 13. ocatios are a ver importat item as these figures have a large effect o how itervetio costs are ultimatel tabulated. Please sped sufficiet time to get ver good estimates of these data. A respose is give for each questio, eve if the respose is "do't kow" or "ot applicable". 3e.Recurret Expeses, Goods Adequate descriptio is give for each idividual item > $1000, to 1 allow market pricig. If the aswer to Q1 is "es", a umber is madator here. 3., 4. If both are 0, Q5 Q9 should be NA. 4. If 0, the Q3 must also be 0 for that item Amouts give i local currec. If ot, a ote should documet the tpe of currec used. Items lie items from records are iserted, begiig at row 14, for all lie items that do't match exactl oe of the items listed i rows A explaatio is provided for a item that is double-couted. Items (This ma occur a time lie items i records subsume idividual items listed i rows 1-10.) Adequate descriptio is give for each lie item, so that we ca Items determie what does ad does ot belog i each categor of recurret expeses. 3., 4., Uits are oted as ecessar (e.g., each uit is a "case of 100", rather tha as a sigle cout). ocatios are a ver importat item as these figures have a large effect o how itervetio costs are ultimatel tabulated. Please sped sufficiet time to get ver good estimates of these data. The aswer is approximatel 12 * Q3, if purchases are doe evel 4. over time If ot, a ote is provided explaiig the uexpected aswer. The aswer is approximatel 12 * Q7, if purchases are doe evel 8. over time. If ot, a ote is provided explaiig the uexpected aswer. 11. If this is more tha zero, the Recurret goods row of Q3 of 3a. Total Facilit Expeditures sheet must also be more tha zero. 1 If 0, Q13 must be NA 15. % allocatios sum to 100%, for each row of data. A respose is give for each questio, eve if the respose is "do't kow" or "ot applicable". 3f.Recurret Expeses, Services A explaatio is provided for a item that is double-couted. (This ma occur a time lie items i records subsume G-5 Versio 2/6/2004 PAN_ma_QA_v4

6 PANCEA Maual idividual items listed i the Service descriptio colum.) 3, 4, 6 The amout give is i local currec. If ot, a ote should documet the tpe of currec used. The aswer is approximatel 12 * Q1, if purchases are costat over the ear. If ot, a ote is provided explaiig the uexpected aswer. 5. % allocatios sum to 100%, for each row of data. 6. If this is more tha zero, the Recurret services row of Q3 of 3a. Total Facilit Expeditures sheet must also be more tha zero. ocatios are a ver importat item as these figures have a large effect o how itervetio costs are ultimatel tabulated. Please sped sufficiet time to get ver good estimates of these data. Items Adequate descriptio is give for each lie item, so that we ca determie the ature of the service ad, if possible, the resources used to deliver it (see maual). Items Spedig for labor is reported either here or i Persoel, but ot i both places. lie items from records are iserted, begiig at row 20, for Items all lie items that do't match exactl oe of the items listed i rows A respose is give for each questio, eve if the respose is "do't kow" or "ot applicable". 4.Fiace 4. Percetages sum to 100% or less. 8. The aswer is > = 0% ad < =100%. 9. Aswers sum to 100%. A respose is give for each questio, eve if the respose is "do't kow" or "ot applicable" a. Iputs -- facilit expeditures Last moth should geerall ot exceed Most recet FY ad usuall will be much smaller Data should cover the etire 12 moths of the most recet FY. This ca be doe either b usig the moth colum or the quarter colum. Last moth should geerall ot exceed Most recet FY ad usuall will be much smaller Data should cover the etire 12 moths of the most recet FY. This ca be doe either b usig the moth colum or the quarter colum. Last moth should geerall ot exceed Most recet FY ad usuall will be much smaller Data should cover the etire 12 moths of the most recet FY. This ca be doe either b usig the moth colum or the quarter colum. For the same perso the Job title should be the same oe used i questio 3 of Persoel roster G-6 Versio 2/6/2004 PAN_ma_QA_v4

7 PANCEA Maual A respose is give for each questio, eve if the respose is "do't kow" or "ot applicable". 5b. Outputs -- facilit expeditures The list of services should iclude those services that costitute the bulk of the facilit s fuctio. Last moth should geerall ot exceed Most recet FY, usuall will be much smaller, ad ma equal about 1/12 as much. Data should cover the etire 12 moths of the most recet FY, if available. This ca be doe either b usig the moth colum or the quarter colum. ARQvct 1. The date give is 1-2 moths prior to the curret date. This must be a 12-moth period. This period should rel o records, if a records exist This date is i agreemet with the date give i the ARQi, facilit characteristics, Q9. If es for Was this the same for ALL of the most recet fiscal ear? the the aswer must be NA for (If "o") What did ou provide previousl? ad for (If "o") Whe did ou make this chage? This rule holds for all questios with this structure. 6, No oe listig should be > tha the global aswer give i the first row of Q6. 6. Me > = MSM; Wome > = Pregat wome Q7 < = Q8, because pre-test couselig is usuall doe i groups. 8. Me > = MSM; Wome > = Pregat wome Q9 < = Q8. Number of tests admiistered will almost alwas be 9, less tha the umber of idividuals receivig pre-test couselig, ad will ever be more. 10. Q10 < Q9 11 Q11 < Q No oe listig should be > tha the global aswer give for Q Me > = MSM; Wome > = Pregat wome 1 Q12 < = Q Q13 > = Q This is oe of two of the most essetial output iformatio items both for the last moth ad for the last fiscal ear b time period. Whatever iformatio is available should be etered. Maximum data collectio effort idicated. G-7 Versio 2/6/2004 PAN_ma_QA_v4

8 PANCEA Maual This is oe of two of the most essetial output iformatio 14. items both for the last moth ad for the last fiscal ear b time period. Whatever iformatio is available should be etered. Maximum data collectio effort idicated 15. This is a importat questio. Take eough time to get a detailed aswer. 20. If oe o oe, the Q7 should = or be ver close to Q7, (i.e., sessios = idividuals) 31. If Q31. > 40 hours, services must be offered o some eveigs or weekeds (Q32). 35. Q35 < = Q36. Could be o the same da if usig rapid tests. 41. If Q42 = 0, Q41 = 0 or NA 42 If a price is listed i Q40., the Q4 ca't be 100%. (I the rare case that this happes, a ote should be provided.) 39. If Q42 = 100%, the Q39. must be "o" or NA. The aswer should be reasoable, cosiderig the umber of 43. couselors listed o the staffig sheet i the ARQi, ad give the umber of cliets served. (Q6.) > = the largest of the followig: Q50, Q51, or Q5 "das" 54 questios. 60. Q5 should list differet facilities tha Q60. ARQ sti 1. The date give is 1-2 moths prior to the curret date. This must be a 12-moth period. This period should rel o records, if a records exist. 3. This date is i agreemet with the date give i the ARQi, facilit characteristics, Q9. If es for Was this the same for ALL of the most recet fiscal ear? the the aswer must be NA for (If "o") What 4. did ou provide previousl? ad for (If "o") Whe did ou make this chage? This rule holds for all questios with this structure. 6, No oe listig should be > tha the global aswer give i the first row of Q6. 6. Me > = MSM; Wome > = Pregat wome Q7 < = Q6. 8. Q8 > = Q6. 8. Me > = MSM; Wome > = Pregat wome 8, No oe listig should be > tha the global aswer give for Q8. G-8 Versio 2/6/2004 PAN_ma_QA_v4

9 PANCEA Maual 9. No oe listig should be > tha the global aswer give for Q9 This is the most essetial output iformatio both for the 10. last moth ad for the last fiscal ear b time period. Whatever iformatio is available should be etered. Maximum data collectio effort idicated. 10. Make sure that data is provided for a full 12 moths i the Services: FY detail sectio. 10. Respective aswers are < = 9 for etiologic diagoses 14. > = HIV listig i Q < = HIV listig i Q This is a importat questio. Take eough time to get a detailed aswer. 23. The aswer for STI services should approximate facilit hours b give i ARQi "facilit characteristics" Q16. If Q23. > 40 hours, services must be offered o some eveigs or 24. weekeds < The prices listed must be less tha the prices give i Q3 If a price is listed i Q35., the Q36. ca't be 100%. (I the rare 36. case that this happes, a ote should be provided.) If prices were iclusive i Q33., the should be iclusive here, ad 36. visa versa. 39. If Q36 = 100%, the Q39. must be "0".or NA 40. If a price is give i Q39., the the aswer here must be < 100%. 41. If Q40. is 100%, the aswer here should be "o". 4 The price give here should be less tha the price give i Q39. The aswer should be reasoable, cosiderig the umber of 43. couselors listed o the staffig sheet i the ARQi, ad give the umber of cliets served. (Q6.) > = the largest of the followig: Q51, Q52, or Q53. "das" 54 questios. 60. Q5 should list differet facilities tha Q60. G-9 Versio 2/6/2004 PAN_ma_QA_v4

STRAND: FINANCE. Unit 3 Loans and Mortgages TEXT. Contents. Section. 3.1 Annual Percentage Rate (APR) 3.2 APR for Repayment of Loans

STRAND: FINANCE. Unit 3 Loans and Mortgages TEXT. Contents. Section. 3.1 Annual Percentage Rate (APR) 3.2 APR for Repayment of Loans CMM Subject Support Strad: FINANCE Uit 3 Loas ad Mortgages: Text m e p STRAND: FINANCE Uit 3 Loas ad Mortgages TEXT Cotets Sectio 3.1 Aual Percetage Rate (APR) 3.2 APR for Repaymet of Loas 3.3 Credit Purchases