Results & exports · Administrator
Read the analysis pages and historical charts
Select the question and round, inspect distributions and statistics, and interpret revisions as events rather than new participants.
Screenshots illustrate the controls below. Layouts and labels may vary with your survey configuration. Select an image or “View full size” to open it in a new tab.
Select the dataset
Open Analyze Survey, choose the relevant question or matrix cell, and select the survey round. Check the language and any filters before comparing results.
| Tab | Purpose |
|---|---|
| Overview | Consensus status, group stability, respondent activity, and changes over answer events. |
| Distribution | Counts or percentages across choices, with available charts and history. |
| Statistics | Enabled numerical measures and their histories. |
| Comments | The panel’s written reasoning. |
| Sentiment / Entities | Optional NLP outputs where the service is enabled. |
More screenshots and examples (1)
Interpret the timeline
A chart event can be a new answer or a revision to an existing answer or comment. Events need not occur at equal time intervals. Ten displayed events do not imply ten people or ten days.
Check the chart’s displayed window and round selection. Historical graphs do not automatically constitute a protocol-defined stability assessment. Use the underlying exported data when you need reproducible calculations.
More screenshots and examples (2)
Written input and optional NLP
Read comments in context. Optional Natural Language Processing can provide sentiment, entity, and phrase outputs when enabled for the account. It is separately licensed and is not a substitute for reviewing the meaning of research responses.
If your project restricts third-party text processing, confirm that NLP is disabled before data collection. Do not describe sentiment scores as consensus or assume the service is included with every account.
More screenshots and examples (3)
Detailed instructions: Analyze Survey
Surveylet provides in-depth analysis for your survey questions. Depending on the question type, different analysis tabs are shown.
To select the question for analysis, hover your mouse over the 
icon to see the list of questions available for analysis in your survey. The numbers in parenthesis in front of each survey question represent the "(Page/Rank)" of the question. For example, "(5/2)" refers to the 2nd question on page 5.
icon to see the list of questions available for analysis in your survey. The numbers in parenthesis in front of each survey question represent the "(Page/Rank)" of the question. For example, "(5/2)" refers to the 2nd question on page 5.
When a matrix question is selected, you can pick the matrix row and matrix column by hovering your mouse over the icons again displayed in front of the matrix row and matrix column lists:
icons again displayed in front of the matrix row and matrix column lists:
Using the dropdown box located in the upper right corner of the page, you can switch to other survey rounds. Each survey round is displayed with its start date. Switching to another survey round will run analysis for the selected question as it was at the end of that survey round.
You can select the languages to display the analytics from the language dropdown in the upper right corner right next to survey round selector. The languages displayed under this dropdown will be languages you set up in Survey Settings > Display Settings > Survey Design Languages option.
Detailed instructions: Comments Tab
The Comments tab displays open text analysis of all answers for My Comments or open text (Essay) questions. There are 3 sections.
Detailed instructions: Distribution Tab
A distribution in statistics is a function that shows the possible values for a variable and how often they occur. The Distribution tab of the Analyze Survey page displays three distribution charts.
Vertical Bar Chart
A bar chart depicts numeric values against distinct categories in a two-axis plot. On one axis, the category levels are listed. For each category, one bar is plotted, and its length along the other axis corresponds with the numeric value associated with the category. The example bar plots below show how a user base is divided among different choices.
Pie Chart
A pie chart shows how some total amount is divided among distinct categories as a circle (the namesake pie) divided into radial slices. Each category is associated with a single slice whose size corresponds with the category’s proportion of the total. The figure below plots the same data as above, but using the pie chart form instead.
Historical Line Chart
A line chart is a type of chart which displays information as a series of data points called 'markers' connected by straight line segments. A line chart is used to visualize a trend in data over intervals of time – a time series – thus the line is drawn chronologically. They are known as run charts. The Line Chart on this page displays the distribution values collected over a period of time. Each point in the x-axis represent an event in time, where the event is generated when a new response is received or an existing answer is revised for a question. The events are not distributed in time equally; they may not happen at the same intervals.
Detailed instructions: Entities Tab
Surveylet's Natural Language Processing (NLP) engine carries out entity and phrase-level extraction and the results are displayed on the Entities tab.
Entity Extraction: Extract the relevant named entities (personal names, places, companies, addresses, dates, phone numbers, etc.). Entity extraction is available in English, French, German, Spanish, Italian, Portuguese, Dutch. Surveylet displays the top 15 most frequently used entities on the Entities screen.
Phrase Extraction: Extract the relevant multi-word noun, verb, adjective or adverbial phrases using morphological and syntactic analysis. Phrase extraction is available in English, French, German, Italian, Spanish, Portuguese, Catalan, Dutch. Surveylet displays the top 50 most frequently used entities on the Entities screen.
Detailed instructions: One Page View
You can display all analytics tabs on a single page by clicking on the print icon at the bottom of the page.You can remove sections you do not need by appending any of the following items to the end of your query string.
Example:
&hidesections=Sentiment,Entities
Application:
https://www.calibrum.net/srvcas1/analyze.asp?x=42528933&rd=116068&ui=Ai1CdGKv%25A1V%2593%258F&si=kkY%252Db%23%D&bchst=CONS_STABILITY&qcti=&lchst=CONS_CONSENSUS&lang=EN&round=0&qi=%2C%1V%2593253Cxaa&print=1&hidesections=Sentiment,Entities
Available list:
Overview
Distribution
Timeline
Self
Sentiment
Entities
Statements
Texts
Keywords
Comments
Statistics
GroupSentiment
FYI: Surveylet does not export survey pages to PDF, but you can use 3rd party browser extensions such as FireShot free of charge to export pages as images or PDF files.
Detailed instructions: Overview Tab
The Overview tab provides you with an overall analysis of the survey, consensus, stability and responses.
Consensus Status
The status of group agreement. Either Consensus or Disagreement.
Group Stability
Stability of Consensus (also known as Strength of Consensus)
Consensus Measure
The statistical measure used to calculate consensus.
Stability Measure
The statistical measure used to calculate Group Stability. Stability Measure settings is ignored for Range Consensus measure; "Range X" is displayed.
People Invited
The number of people invited to the survey. This is the number of people who are member of the survey participants group (i.e. the Surveylet Group your survey is assigned to).
Survey Participants
The number of panelists who clicked on your survey invitation link and entered the survey at least once.
Responses
The number of people who answered this question.
No Response
The number of people who have not yet answered this question.
Comments
The number of comments this question has received.
Revisions
The number of total revisions of answers/comments for this question.
This vertical bar chart represents the responses and revisions made by panelists. Each bar is assigned to a panelists. The height of the bar indicates how many times the respondent answered/revised his/her answer to this question. If the Answer is 1, it means that the panelist answered the question only once and did not revise it afterwards. If for example the Answer is 10 (i.e. the height of the orange bar is at 10, it means that the panelist revised his/her opinion 9 times after s/he answered it). If you hover your mouse over a vertical bar, you will see response details, such as name (email), number of answers, and the date time of the last answer. If the survey's Privacy Settings is set to High, you as the administrator will not be able to see the name (email) of the panelist to hide identity.
This historical chart displays the evolution of consensus. The blue line is the Consensus line graph (also known as line chart or running chart) displaying Consensus trend. Each green/red bar represents an event in time (either a new responses event, or a revision of existing response event). Events may not be of the same distance from each other in time as these events may happen randomly and not periodically. During a 2 week survey, maybe 80% of responses and revisions are submitted within the last few hours. The red horizontal line across the 50% mark represents the Stability Threshold.
The green vertical bars indicates the successful status of consensus at that time; the red vertical bars indicate that at that moment there were disagreement among panelists regarding this question. The 0 - 100 vertical coordinate on the left side is the Stability Status from 0% - 100%. The 1 - 5 values vertical coordinate represents the ranks of the 5 choices of that are assigned to this question (1 - Strongly Disagree, 2 - Disagree, 3 - Neutral, 4 - Agree, 5 - Strongly Agree). The chart display the 50 most recent events. The x coordinate at the bottom represents the number of events starting from 1 to 50.
Detailed instructions: Sentiment Tab
Sentiment analysis refers to the use of natural language processing, text analysis, computational linguistics, and biometrics to systematically identify, extract, quantify, and study affective states and subjective information. Sentiment analysis lets you analyze the sentiment behind a given piece of text. In this article, we will look at how it works along with a few practical applications.
Sentiment analysis is a technique through which you can analyze a piece of text to determine the sentiment behind it. It combines machine learning and natural language processing (NLP) to achieve this. Using basic Sentiment analysis, a program can understand whether the sentiment behind a piece of text is positive, negative, or neutral. It is a powerful technique in Artificial intelligence that has important business applications. For example, you can use sentiment analysis to analyze customer feedback. After collecting that feedback through various mediums like Twitter and Facebook, you can run sentiment analysis algorithms on those text snippets to understand your customers' attitude towards your product.
Sentiment analysis is an optional service and it requires the purchase of the Natural Language Processing services. It uses a combination of NLP and machine learning algorithms. Surveylet Sentiment Analysis service analyzes opinions found in texts. It first identifies the topic(s) (topic=entity) that are being discussed in a particular text and then evaluates the opinion(s) expressed about the topic(s). The result is a numeric score that is either positive or negative.
For example, for the sentence “I have a beautiful house in London, but it is in a very bad neighborhood.” Surveylet sentiment analysis would identify the topic being discussed as ‘house’ and then establish a relationship between 'house' and 'beautiful’ based on the Surveylet engine’s deep knowledge of language. ‘Beautiful' is understood as a positive opinion expressed about the topic ‘house’ and so the numeric score of the sentiment analysis would be positive. A relationship is also established between 'neighborhood' and 'house' whose sentiment is negative. When a text is sent to the sentiment analysis API endpoint, it is divided into different sentences, and analyses are automatically performed on each sentence. If a sentence contains no sentiment expression, it receives a score of ’0’ and the corresponding fields for ‘topics’ and ‘sentiment expressions’ are left blank. Surveylet combines the positive and negative entities in each sentence to arrive at a sum of total value which will be the score of the sentence.
The Natural Language Processing module is using a linguistic analysis for entity extraction, phrase extraction and entity analysis (see below).
-
Sentiment Analysis: Identify the topics of conversation and evaluate, extract and quantify the emotional state or attitude of your customers towards those topics with a polarity score value.
-
Phrase Extraction: Extract the relevant multi-word noun, verb, adjective or adverbial phrases using morphological and syntactic analysis.
-
Entity Extraction: Extract the relevant named entities (personal names, places, companies, addresses, dates, phone numbers, etc.).