I am a developer and I have a problem trying to design a system to
manage data coming from web surveys. Each section can potentially have
dozens of questions, i.e., fields.
I am focusing here only on the table(s) that will hold the survey data.
I do not have any DDL as I am still trying to understand this!
All the examples I have found so far in books and on the web seem to
deal with fairly limited data, that is easily, or so it looks, broken
down in multiple tables.
It seems that, from my research, having a wide table per survey section
with each field as a column, which has been suggested to me, is not
proper design for many reasons - missing values for non-required
questions, table with 100s of possible columns, etc... - so I played
with the idea of a single table where one of the columns would be the
foreign key pointing to a questions table and another column would hold
the data (this is a simplified explanation.)
The problem with this is that now this column will have to accomodate
all types of data, from bits to large varchars, and that field
validation seems now impossible jeopardizing data integrity.
What do knowledgeable designers come up with in this case? Can someone
point me in the right direction?
I've written several survey systems in which the majority of the questionshave the same or similar responses (Yes/No, True/False, scale of 1 - 5,etc).But this latest survey system I'm working on has 8-10 sections, with avariety of question attributes and answer scales. Some items have just adescription and require a Yes/No answer, others have a description and anactive status and require a Yes/No and price answer, some require a comment,etc.Rather than build a separate response table for each survey section, I wasthinking of building one generic response table, and trying to force allsections to fit by adding columns - some of which won't apply to some items.Like this:Survey Category (will apply to all items)Survey Section (will apply to all items)Item Description (will apply to all items)Item YN (will apply to all items)Item Price (will apply to about 10% of the items)Item Points (will apply to about 10% of the items)Item Active YN (will apply to about 10% of the items)Item Fail YN (will apply to about 10% of the items)Item Comment (will apply to about 10% of the items)For instance, in the structure above the field "Item YN" would representmultiple types of answers: is the item in use?, is the item in place?, isthe item given away for free?, is the item on display?, etc. Basically,anywhere a Yes/No answer is used.The advantage is one source table (rather than 8) for storing answers, andit might be easier to query and report on.The disadvantages I see are 1) it's more difficult to understand the meaningof the responses when the answer field is named Item YN, and 2) you have anon-normalized table that's difficult for a 3rd party to understand.If I have the questions and responses in separate tables, I'll use nameslike "ItemComplimentaryYN" and "ItemUsedYN" depending on the question. It'seasier for others to learn the data.I actually don't like the "generic" approach, and probably won't use it, butI figured I'd try to get some input from others who've written surveysystems.Thanks
I am a developer and I have a problem trying to design a system to manage data coming from web surveys. On average each subject will take part in more than one survey and each survey may potentially have 100+ questions.
I am focusing here only on the table(s) that will hold the survey data.
I'm thinking about having a wide table storing each question in the survey(s) as a column against the subjects. My main concerns with this approach:
- the "yuckiness" and potentially other performance issues(?) associated with a table with 100s of columns - the 8kb size limit per row (unlikely to touch it, but possible) - 1024 column limit (unlikely to touch it, but possible)
Another approach obviously is to have a single table where we have, in a simplified version, 3 columns: person_id, question_id and data/ answer. The problem with this is that the "data" column will have to accommodate all types of data, from bits to varchars, and that field validation seems now impossible jeopardizing data integrity, and most importantly, you can't easily work with the data in filtering/ reporting etc.
What do knowledgeable designers come up with in this case? Can someone point me in the right direction?
SQL Server Integration Services team would like to hear from you. Our goal is to understand your current Data Extract, Transform & Load (ETL) practices and prioritize features for future releases.
You will have a chance to win one of five 80GB Zunes. To win a Zune, you must fill out the survey by Dec. 20, 2007.
To participate in the survey, please click here: https://mscuillume.smdisp.net/Collector/Survey.ashx?Name=SqlETLSurvey2
Hi,We are a systems research group at the Computer Science departmentatRutgers University, and are conducting a survey to understand detailsabout network, systems and database administration. We hope that thisinformation would help us recreate a realistic environment to helpresearch in 'systems management'.We request network, systems, and database administrators to take thissurvey. As an incentive, all surveys completed in their entirety willbeentered into a drawing of a number of $50 gift certificates (fromAmazon.com).We hope you have a few minutes to take the survey which is located at:http://vivo.cs.rutgers.edu/administration_survey.htmlResearch in our group:The goal of our research is to improve the overall availability andmaintainability of services. Since administrators form an integral partof these services, a key aspect of this work is to build environmentsand tools that ease the task of service administration. In particular,environments which would help administrators know how their actionsmight impact the real service (before performing them for real), webelieve would be useful in preventing inadvertent actions. This surveytries to understand the existing environments, what administrators docurrently to test the 'validity' of their actions, and the difficultiesthey face in doing so. The two specific systems we are looking at arenetworks and databases, as we believe these are important components ofmany services.If you have any questions regarding this survey or our work, feel freeto email us:Fabio Oliveira (fabiool at cs || rutgers || edu), orKiran Nagaraja (knagaraj at cs || rutgers || edu)Thanks for your time,Fabio OliveiraPhD student, Vivo Research Group (http://vivo.cs.rutgers.edu)Rutgers University
Dear friendsI am conducting a survey on Relational Database usage and would likeyour help. The study is part of my MBA Dissertation.Could you kindly spare 5 minutes to take part in this survey?http://FreeOnlineSurveys.com/rendersurvey.asp?id=120816ThanksRajeev
Ok, I'm doing a football database for fixtures and stuff. The problem I am having is that in a fixture, there is both a home, and an away team. The tables as a result are something like this:
It's not exactly like that, but you get the point. The question is, can I do a fixture query which results in one record per fixture, showing both teams details. The first in a hometeam field and the second in an away team field.
Fixture contains the details about the fixture like date and fixture id and has it been played
Team contains team info like team id, name, associated graphic
TeamFixture is the table which links the fixture to it's home and away team.
TeamFixture exists to prevent a many to many type relationship.
Make sense? Sorry if this turns out to be really easy, just can't get my head around it at the mo!
I am trying to modify a survey DB schema created by OOP(Object Oriented Programming) snobs who had their heads so far up their butts(in my opinion). This is current schema created by them.
Actually there aren't a several foreign keys as they are shown in the picture, but that's what they(OOP snobs) are doing to the table "Response". They are inserting question_collection_id and its name as well as question_id and its name to Response table. However, it was not ill-intended, "We just wanted to load the report quickly"(OOP snobs). But I really recommend this schema, Because of its simplicity.
Yet I haven't proven that my schema will provide report just as quickly as OOP snobs' version. As mentioned earlier, since they don't have foreign key constraints accordingly, I do find quite number of discrepancy on values of "question" columns between "Question" and "Response" tables and "question_collection" columns between "question_collection" and "Response" table. I can provide schedule task to eliminate those problems. But I really prefer my schema than theirs. So far, I can explain mine is better just because of its simplicity and DBA's gut instinct. I'd like to hear Superior Expert's opinion which schema is better and why one is better than the other.
Attached is also zip files of schema images if you can't see them by those URL.
The BI Survey is the largest independent survey of BI technology users worldwide. (It used to be known as the OLAP survey, but has expanded to cover a broader array of BI technologies.) The survey analyzes people's buying decisions, BI implementation cycles and business results achieved.
The BI Survey is strictly independent --- they ask vendors like us to ask our customers like you to participate, but we don't have any other involvement --- except closely reading the results, of course. The Survey will send all participants a summary of the full survey results and place them in a drawing for ten $50 Amazon vouchers.
You can complete the survey on-line by following these links:
English version - website: http://www.eu-survey.com/webprod/cgi-bin/askiaext.dll?Action=StartSurvey&T1=BI&lan=1&lin=20
German version: http://www.eu-survey.com/webprod/cgi-bin/askiaext.dll?Action=StartSurvey&T1=BI&lan=2&lin=67
I am going into salary negotiationyearly review next month after my week off and I am trying to determine what to ask for. I am thinking another 10K and an extra week off would not be unreasonable, but I wanted to get some idea of what you guys think I should ask for in terms of pay.I have been developing software for seven years and I have been a dba for a little over five. I live in pricey northern Virginia. On a fairly regular basis I do about 55 to 60 hour weeks. On my DBA team, I am the only one who can handle both development and production tasks. The others are strictly developers. Although I have been relieved of most of my customer support tasks by our newbie, the customer support manager still brings the nastier bits to me. It is my perception that the more complex tasks get assigned to me. I get told on a regular basis that I am the best dba this place has ever had and other embarrassing accolades are regularly thrown my way. After a year, other than my boss I am the dba that has been here the longest in high turnover high burnout company. This year as we try to move to a SAP model, it looks like we will be going to 24/7 support on a disaster recovery model I am designing and implementing, so I guess I am getting the pager.So how much money should I be asking for?you can PM me with a number if you want.
I have surveys that I need to add weights to and was wondering if there was a way to convert the contents of a column.
select empid, ans_for_ql, (if ans_for_ql = A then 0, B then 3, C then 5) as weightscore, ans_for_q2, (if ans_for_q2 = A then 8, B then 4, C then 2, D then 1, E then 0) as weightscore
Here is what the table looks like:
empid | Ans_for_Q1 | Ans_for_Q2 1001 A C 1002 B E
And these are the possible answers and what they need to be converted to:
Weights for answers to Question1 Q1_A = 0 Q1_B = 3 Q1_C = 5
Weights for answers to Question2 Q2_A = 8 Q2_B = 4 Q2_C = 2 Q2_D = 1 Q2_E = 0
from (select distinct(clientID) ,sum(case when QuestionID = 1 and Answer in (1,2,3,4,5,6,7,8,9,10) then Answer else 0 end) as Q1_1 ,sum(case when QuestionID = 1 and Answer in (1,2,3,4,5,6,7,8,9,10) then Answer else 0 end) as Q1_2 ,sum(case when QuestionID = 1 and Answer in (1,2,3,4,5,6,7,8,9,10) then Answer else 0 end) as Q1_3 ,sum(case when QuestionID = 1 and Answer in (1,2,3,4,5,6,7,8,9,10) then Answer else 0 end) as Q1_4
from dbo.survey
group by clientid) as x
group by x.clientID ,x.Q1_1 ,x.Q1_2 ,x.Q1_3 ,x.Q1_4
Dear Madam/Sir,You are invited to participate in an international research study. Thisresearch project is headed and led by Cambridge student Nico Baumgarten.What is it all about?It is a well-known fact that there exist many differences between cultures.However, how these differences effect motivation if there are effects atall is not yet clear. This survey is set out to change this. With yourhelp, it will be possible to achieve this.Therefore, we kindly ask you to spare only 10 minutes of your time andvisit the survey home page.Go to the survey http://crosscultural.research.yi.orgor if this does not work the direct link http://82.4.146.32Who participates?All over the world, people are chosen to participate in the survey.Although the survey is completely anonymous, each person is arepresentative of his/her country. Hence, it does not matter, which job aperson has, whether he/she is male or female, old or young. The more peoplechoose to participate the more accurate this survey is going to representand show cultural influences as well as similarities.So if you want to tell your friends about this survey - Please Do So! Justforward this email to your friends or use the feature on the survey homepage.What is in it for you?With your help your country will be represented in the survey.Of course, you can opt-in to receive the results of the study at the end ofthe survey.Any information you provide will be kept strictly confidential andanonymous.Thanking you in advance for you support I remain withKind regardsNico Baumgartenhttp://crosscultural.research.yi.orghttp://82.4.146.32
I am working on a survey system with another DB Engineer.Our current design includes 1 table per Question Type (Numeric, Boolean, Text), Response values are tightly coupled based on data being entered, to keep things simple assume there are only 6 tables in this this system (1 for each Question Type to define the "rules" of the data entered, and 1 for each Response collection).He is a former .Net programmer and has been talking with the .Net guys on the project and he is proposing that we abstract our Question Types to use one table and store the Input Values all as nvarchar(max).This obviously simplifies the DB design and reduces the amount of work I will need to do so I should be all for it. I guess I am a little concerned that we are no longer tightly coupling the data types and essentially leaving all the validation up to the application. Also should I be worried that a bunch of bit data will be stored as 1/0 in an nvarchar(max) column?Thanks,-John
Several prominent researchers are conducting a survey and looking for500 respondents by Friday on the topic of enterprise architecture andsoftware development. Here is the link to the survey:http://www.surveymonkey.com/s.asp?u=845572436515After completing the survey, results are displayed immediately. A smalldonation will be made to a charitable organization after itscompletion.http://duckdown.blogspot.com/
- Sales Volume: Information about sold articles to a customer incl. Selling date - Survey: irregular answered survey questions about customers incl. date of answer and three Dimensions: - Customer - Date - Survey Answer: Information about possible Answer values (e.g. Yes / No)
We would like to be able to determine the aggregated sales volume (sum) of a customer for a specific period depending on the latest survey answer within this period.
For example: Selected Time period: Jan - Jul 2015 Sales Volume Customer X - Jan - Jul 2015: 1000 Litres Sales Volume Customer Y - Jan - Jul 2015: 500 Litres
Surveys answered:
15th Jan 15: Customer X, Survey Question A: Yes 2nd Mar 15: Customer X, Survey Question A: No 20th Apr 15: Customer X, Survey Question A: Yes
10th Feb 15: Customer Y, Survey Question A: Yes 20th Jul 15: Customer Y, Survey Question A: No
Latest survey answer (Jan-Jul) Customer X, Question A: Yes Latest survey answer (Jan-Jul) Customer Y, Question A: No
Excel Pivot should show something like this:
Question | Latest Answer: Yes | No |
------------------------------------------------------------------------------------ A | 1000 Litres | 500 Litres |
I actually work in an organisation and we have to find a solution about the data consistancy in the database. our partners use to send details to the organisation and inserted directly in the database, so we want to create a new database as a buffer database to insert informations from the partners then make an update to the main database. is there a better solution instead of that?
Hello everyone,I have a webcontrol that uses database-structures alot, it uses the system tables in SQL to read column information from tables. To ease the load of the SQL server I have a property that stores this information in a cache and everything works fine.I am doing some research to find if there are anyway to get information from the SQL server that the structure from a table has changed.I want to know if a column or table has changed any values, like datatype, name, properties, etc.Any suggestions out there ?!
I have a system that basically stores a database within a database (I'msure lots have you have done this before in some form or another).At the end of the day, I'm storing the actual data generically in acolumn of type nvarchar(4000), but I want to add support for unlimitedtext. I want to do this in a smart fashion. Right now I am leaningtowards putting 2 nullable Value fields:ValueLong ntext nullableValueShort nvarchar(4000) nullableand dynamically storing the info in one or the other depending on thesize. ASP.NET does this exact very thing in it's Session State model;look at the ASPStateTempSessions table. This table has both aSessionItemShort of type varbinary (7000) and a SessionItemLong of typeImage.My question is, is it better to user varbinary (7000) and Image? I'mthinking maybe I should go down this path, simply because ASP.NET does,but I don't really know why. Does anyone know what would be the benifitof using varbinary and Image datatypes? If it's just to allow saving ofbinary data, then I don't really need that right now (and I don't thinkASP.NET does either). Are there any other reasons?thanks,dave
Hi All,Can u please suggest me some books for relational database design ordatabase modelling(Knowledgeable yet simple) i.e. from which we couldlearn database relationships(one to many,many to oneetc.....),building ER diagrams,proper usage of ER diagrams in ourdatabase(Primary key foreign key relations),designing smallmodules,relating tables and everything that relates about databasedesign....Coz I think database design is the crucial part of databaseand we must know the design part very first before starting up withdatabases.....Thanks and very grateful to all of you....Vikas
Hi All, I am designing database where few of the master tables will reside in different database or in case different server. Scenario is Server "A" with Database "A" may host the "Accounts" table. Server "B" with Database "B" may host the "Product" table. I am designing database "Project" which will hosted in Server "A". My application requires this master tables [readonly access] as data inserted in my application refers this tables. Also there are reports to be generated which refer this tables. How do i design my database and sql queries? I am thinking of approach of having equivalent tables created in my database and writing service which keep tables in my database in sync. This will ensure good perfomance during transaction and reports as they will need to refer this table locally as opposed to different database or different server.
Any thoughts on above approach?? or any better/standard way for such scenarios ?
Thanks in Advance. Your inputs will be of great help.
Online US Searchable Map of the 50 US States. Users search criteria is the following: Query records by selecting state, county, then record. Each County table has 10-20 tables. All databases combined = 500MB and TLogs = 100MB.
How would you re-design a relational DB where users could query data by state-county-record. Currenty the DB's are created by the County of each state which creates hundreds of DB's in SQLServer with no realtionship to each US state. What would be the best design to ensure good performance, data integrity and maintenance? Would you create 1 DB with all 50 states, create 4 DB's and divide by region(N,S,E,W), 50 DB's of each state or leave it as is with each county it's on DB? Any suggestions would be appreciated.
Hello, I am designing my first database with 5 tables for a demo project and am not sure if it works. an example below.2 of the many things I want visitors to the site to do is find a company by the industry sector they belong to,..andwhat sort of service or products they can supply. For instance a Employment agency maybe under professional services Table 1 Customer Customer_ID = primary key,,,, Sector_ID = Foreign keyComapany Name, Address, Phone, Postcode etcTabel 2 Industry SectorsSector_ID = primary key,,,,Customer_ID= foreign key banking, Education,Prof Services, etc Table 3 Trading ActivityTrading_ID = primary key,,,,Sector_ID = Foreign key, Products_ID= FkEmployment Agent, School, Lawyer etcTable 4 ProductsProducts_ID = primary key,,,,Trading_ID = foreign keySupply frozen foods, transport services, sports goods, etc Table 5 Account Account_ID = primary key,,,,Customer_ID = foreign keyAccount Name, Credit Limit, Payment Terms, Open date, Account contact etc One big point of confusion is, can I have the Customer_ID from the principal Customers tablein every table as a foreign key or must the tables be chained together one after the other as such. Advice appreciatedThanks
Hi, I need a hand with designing a database. I am collecting results from a survey which has the following questions: Call ref? How did you place your support call? Were you satisfied with the amount of time you had to wait until getting acknowledgement of the support call placed? 1 = very satisfied and 10 = very unsatisfied. How happy were you with the customer service you received upon placing the support call? 1 = very unhappy and 10 = very happy.How satisfied were you with the amount of time you had to wait until you heard from an engineer? 1 = very satisfied and 10 = very unsatisfied. How satisfied were you with the time taken to get your problem/query resolved? 1 = very satisfied and 10 = very unsatisfied Did you feel the engineer had enough knowledge to deal with your call? 1 = very good and 10 = not very good Overall how satisfied were you with the support call placed? 1 = very satisfied and 10 = very unsatisfiedIs there anything we can do to improve the quality of the support and service you received? I want to store this in a database. Obviously I want to use best practice for design, normalisation etc. The stumbling block I am coming accross is the fact that each question has a number and each question has a score from 1 to 10 and storing this in the database. Any help appreciated! Thanks Andrew
I am creating database tables for company testimonials. Database columns: name, position, companyname, comment, service we provided. My question is that for each company - may have a multitude of different services from us, and different people with different positions in the same company may make comments. What is best practice for putting this db structure together? Thanks Andrew
Can anyone tell me how can i design database architecture for the Table Category & Product...so that i can make N-level entity relation....I have database in SQL SERVER 2000.
I guess I am confused about something and need some help. I am looking at a database schema for about 20 tables in a database. I noticed that the firstcolumn in each table Is some type of Id. For example StudentId,TestId etc.Where the Id is a unique numeric sequential value. So I have some questions?1- Do these Id's act as what are called indexes for the table?(Unique indexes)2- If the answer to 1 is correct, then how do I create these unique indexes? Is it as simple as declaring the Id column as the primary key and that this value will be generated automatically upon insert? 3- Is it necessary to have an Id column for every table, or may I only do it for a few of them?4. In relation to Question2, what do I need to do, so that the Id column will automatically be created when someone inserts a value into the database table with an ID column. For example here are two sample tables StudentsInfo StudentId<PK> Name Age 1 Mark 33 2 Jill 23 3 Mary 25 PersonalInfo Name<FK> SocialSecurity Address MajorMark 324-444-3342 15 Elm ArtJill 888-888-8998 21 North ScienceMary 876-777-2344 18 Byle Music
Hi,I am planning to create a technical forum for our college. could any one give me an idea of how to design the database for the forum. what table and columns i should have?
I have got a design issue.I have got 4 tables,having relationships.Now,user will edit records and save them,but I want to save the values which were exixsting before they were edited and saved.Implying,if a record is edited 50 times,then the values of each edit needs to be saved.How best ,can I achieve this.
Hi !!We are designing a system where we ask people for their interests and store in into the database and send customize email. Following are the questions:1) Should we use Identity column as Primary Key and CustomerID column? OR we should create Custom CustomerID and use it as Primary Key? (I have read few articles about Identity column as Primary or not Primary, but need little advice what to accept)2) We have a Tables called : Interest & Customer_InterestCustomer Table:CustomerID, Customer Name, Address, Email, Signup DateInterest Table:InterestID, InterestNameCustomer_Interest: (Need suggestion for How to design this)Should Table be design like:Option1: CustomerID, InterestIDOption2: CustomerID, Interest1, Interest2, Interest3, Interest4i.e.Lets Say: Customer table has CustomerA, CustomerB, CustomerCInterest table has Interest I1, I2, I3, I4Lets Say CustomerA Signedup for Interest I1, I2, I3 and CustomerB signed up I1, I4As per Option1:Customer_Interest Table witll haveCustomerA, I1CustomerA, I2CustomerA, I3CustomerB, I1CustomerB, I4ORAs per Option2Customer_Interest (Where Interest Column is bit column..... 1 = Signed up, 0 = Not Signed upCustomerA, 1, 1, 1, 0CustomerB, 1, 0, 0, 1Which way we should design? 3) If we select Option2, and if we are displaying data in ASP.NET Page, will there be any issue if we use 3 tier architecture?Thanks !!!
