How to mark question number and complete question statement in the survey data?
1) Theoretically, each column in the raw data must contain three elements:
Question number
Question statement
Question attributes (if available)
2) Why must customers provide these information?
When running the report we need these information to setup the report in our system.
However, these information may be missing or incomplete in the raw data downloaded from survey system.
In addition, there are times when customers need to add columns or delete columns, to recode and create new columns, and to add external data, etc.
These tasks may change the original order of the questions.
It is important to double check and complete the above three elements in the raw data.
Here is an example of a multiple choice question. The question in the questionnaire is as following:
The question statement may be missing by some system as shown in the following data example:
London Fair |
Manchester Fair |
Liverpool Fair |
Yes |
Yes |
|
How to complete the information in the raw data?
The Raw data of following example must be in Microsoft Excel file. If your data is in txt or other format you can import into Excel or leave it to us to sort it out.
Insert a blank row above the first row
Input question number for each column, e.g. Q1, or 1. Mark those fields to be used as breakdown fields in the formats like B1, B2, etc.
Tip: For those questions that occupy multiple columns, you can mark the first column of that question only.
See the example below:
2 |
|
|
Which of the following events did you attend? London Fair |
Which of the following events did you attend? Manchester Fair |
Which of the following events did you attend? Liverpool Fair |
Yes |
Yes |
|
Tip: If the original title of a question doesn’t contain the main question statement, input the question statement after the question number. Alternatively, you can insert second blank row and copy the statement in the second row. See the example below:
2. Which of the following events did you attend? |
|
|
London Fair |
Manchester Fair |
Liverpool Fair |
Yes |
Yes |
|
More information for advanced users:
1. You can add external data as breakdown field, please set the question number in the format of “B + number” as B1, B2, etc. See the example below:
B1 |
B2 |
2. Which of the following events did you attend? |
|
|
Age |
Gender |
London Fair |
Manchester Fair |
Liverpool Fair |
25-35 |
Female |
Y |
Y |
|
2. You can combine external data and create report for it. Please mark question number, main question statement and attributes like an ordinary question in the raw data. See sample question 101 in the raw data below:
B1 |
B2 |
2. Which of the following events did you attend? |
|
|
101 |
Age |
Gender |
London Fair |
Manchester Fair |
Liverpool Fair |
Where do you live? |
25-35 |
Female |
Y |
Y |
|
Manchester |
36-45 |
Male |
Y |
|
|
London |
3. You can generate a new question by recoding the values of an existing question. Please mark question number, main question statement and attributes like an ordinary question in the raw data. See sample question 1001 in the raw data below:
B1 |
B2 |
2. Which of the following events did you attend? |
|
|
10 |
1001 |
Age |
Gender |
London Fair |
Manchester Fair |
Liverpool Fair |
Where do you live? |
Where do you live? Recode |
25-35 |
Female |
Y |
Y |
|
Manchester |
Out of London |
36-45 |
Male |
Y |
|
|
London |
London |
4. By default, all the questions in the raw data are included in the report. If you need to exclude some questions from the report, please delete the relevant columns from the raw data directly.
5. Summary and suggestions of marking question number:
Mark number for original questions:
We suggest the question number is marked the same as the original question number in the questionnaire.
Mark number for breakdown filed:
The number for breakdownfield must be in the format of B + number like B1, B2.
Mark number for derived questions:
We suggest adding additional digits on the question number.
For example, in the case of the question number 6, if you need to create a new question based on it, the new question number can be labelled as 601, 602, etc. or as 6001, 6002, etc.
This would make it easy to recognize, in the future, that this question is derived from the question 6.
Mark number for external data:
Similar to the derived questions, We suggest marking three or four digits number for the question. This would be helpful when you need to identify it during analysis. You can define detailed rules for your own convenience, but the quesiton numbers must be integer.
How does InsightDolphin's crosstab system clean the survey data?
This document introduces how the Data Cleansing System works and why it is important. There are reasons listed here but not limited to these:
1. The "N/A" value and its alike, such as "Unknown", "Not sure", "Unsure", etc., are useful in the questionnaire design. But they might need to be deleted from the reports because they enlarge the total number of responses.
2. Your raw data may contain blank values that are not blank actually. These unseen non-blank values must be removed because they silently distort the analysis result.
3. Leading and trailing spaces must be cleared to keep the values consistent.
4. Tidy headings make analysis and reporting job easy.
Below is an example of Before-After Comparison for a single choice question:
Below is an example of Before-After Comparison for a multiple choice question:
Further details:
The quality of survey data are affected by many factors, such as the entry of survey takers,
and the design of back end and front end of various survey platforms. We have developed a data cleansing system,
and have designed a standard process to make sure the data can be used efficiently for data analysis.
This system is used to provide AppenixPro reporting service. In the final report, our client would have a professionally cleaned raw data.
During the data cleansing process, the unseen blank characters, leading and trailing spaces and more will be cleared.
The “N/A” and similar values can also be removed based on the clients' instruction. We ensure that our clients have a
ready to use raw data for further analysis.
Below are some common objects to be cleaned:
How does the final cross tabulation report look like and how to interpret the crosstab report?
Customers would receive an Excel file with suffix of XLSX.
Each question will be placed in a separate worksheet.
A clean raw data is also placed in a worksheet.
Our data cleansing process is systematic and professional.
Using this data will be more reliable and convenient than your original data.
The layout of report is consistent and stable. This makes the researchers, who are familiar with the data tables' structure, focus on the data interpretation and find out results efficiently.
The simple structure is that the values for all breakdown fields are located in the first row, and all attributes are in the first column.
Note that the breakdown column that has no responses do not appear in the tables.
This means that if these columns are added into the tables, their values would be zero or blank.
The way of the report interpretation is almost the same for different types of question.
The following examples show how to read the data table of report:
1) Single Choice question type: the following table is the final report of a sample question Q1
2) multiple choice question type: the following table is the final report of a sample question Q2.
3) rating scale question type: the following table is the final report for a sample question Q3.
The values can be numeric score (like ranking ‘1, 2, 3, 4, 5’) or semantic rating (like rating ‘Strongly agree, Agree, Neutral, Disagree, Strongly disagree’).
You can see that customer can decide the scale group. In this case the group name is “Excellent+Good”.
You can name the group as “Top2” or other names when filling the “Report Requirement Form”.
4) For a question that needs to be calculated for an average
Summary of the layout
Brearkdown values are in the first row
Attributes are in the first column
Breakdown values that has no responses are removed from the table
Grouping for rating questions is unlimited. The values can be either like ‘1, 2, 3, 4, 5’ or like ‘Strongly agree, Agree, Neutral, Disagree, Strongly disagree’.
What is Report Requirement Form and where to download it?
Report Requirement Form is a simple form that a customer must fill in. It indicates the customer's requirements of the crosstab report. By using this form, customers show us which fields are going to be the breakdown fields, and how to group the scales of rating type questions.
Please refer to the instruction link below for details.
After placing order, you will receive an email with Report Requirement Form attached. Alternatively, you can download this form by clicking the link below, or login to your account and download from the link in the Crosstab page.
How to fill in the Report Requirement Form?
1. At the top of the form, there are three check boxes.
The first checkbox is about removing the short value of ‘n/a’, ‘n.a.’, etc. The second one is about removing any value that plays the same roll of n/a, such as Don’t know, not available, etc.
This does not apply to the values that are meaningful choices. In the sample question below, the two choices of “I know” and “don’t know” are equivalent to “Yes” and “No”. So the meaning of “don’t know” is not the same with ‘n/a’ and it would not be removed from the raw data.
5. Do you know how to use Report Requirement Form?
I know
Don’t know
By default the first two checkboxes are ticked. We recommend leaving them as default if you are not sure.
The third check box means that the customer only needs a data cleansing service. If you tick it, please ignore all the rest tables in the form. (Note that the price is the same with crosstab reporting service)
2. The first table is about breakdown field. The gray area is the example area. The breakdown fields are prefixed with letter ‘B’. If you leave the table blank, the final crosstab report contains summary data tables only.
3. The second table is for rating type questions only. You can group the scales using the simple equation: Scale1 + Scale2 = Group Name.
The left of the equal sign are the scale values like ‘Very good’ and ‘Good’. The right of the equal sign is the group name created by users. This group name will display in the data tables of the final crosstab report.
4. The ‘Optional Part 1’ set up the numerical fields to calculated average figures.
Please fill in the column letter from the excel sheet instead of the question number.
5. The ‘Optional Part 2’ is used to exclude some breakdown values from the breakdown field in the report.
Please
click here to download empty Report Requirement Form.
How to pay?
We accept bank transfer. Service is started as soon as the payment arrives to our bank account. This normally takes around three days.
What is Cookie and how is it used?
1) Introduction of cookie
A cookie is a small file downloaded on to your computer when you access a website. A cookie does not contain or collect information.
Cookie is well known and safe. For details about cookie, please go to the following links
https://www.aboutcookies.org
http://www.allaboutcookies.org
2) Use of cookie by our website
According to GDPR, all website must display the consent notice.
We only use cookies to recognize the source of media that customer comes from. The purpose is to help customers get discount.
We do not track any other customer information, such as IP address, user name and password.
We do not use any information for customer analysis.
We do not use the information for advertising purpose.
We do not share the information to any third party.
What is Contingency Table?
Contingency table also know as crosstabulation or crosstab. It shows you number or percentage of cases in the categories of your respondents.
You can find plenty of information on the internet through Google Search and Wikipedia.