Free online questionnaires: from design to analysis

1. 1

   Prepare your items

   Bring in each dimension's items from your own questionnaire, or take one of the ready-made standard questionnaires; signing up is free and needs no bank card.

2. 2

   Lay out scales and matrices

   Put items that share a scale into one Likert matrix, and use single choice, number and province-and-city fields for demographic questions.

3. 3

   Consent and screening

   Put the informed-consent text on the first page, and use conditional logic to end the questionnaire with a suitable message for anyone outside your target population.

4. 4

   Share the link

   Send the short link or QR code on Telegram, Eitaa, Bale or in class, and turn on “One response per person”.

5. 5

   Export and analyze

   Get the responses straight as an SPSS or Excel file and move on to reliability and hypothesis testing.

Most thesis questionnaires have several items measured on a shared scale, from “strongly disagree” to “strongly agree.” The “single-choice matrix” field is the best fit: each row is an item and each column is a point on the scale. On mobile, each row turns into a vertical card so respondents never have to scroll sideways.

For single items you also have numeric scales, stars, sliders and NPS. If option order might affect answers, randomize the order of options or pages and keep options like “None of the above” pinned last.

With conditional logic, questions open only for the people they apply to; work-history questions only for those employed, for example. Screening works too: if a respondent isn't in your target population, the form ends right there with a message of your choosing and their response stays out of the main count.

Most theses set their sample size with Cochran's formula or the Morgan table. For a large or unknown population, at 95% confidence and a 0.05 margin of error, the figure is about 384 people, and it drops for a finite population. Work out the exact number for your own population with theSample size calculator (Cochran's formula).

Many form builders cap the free plan at around 100 responses — so right when you're halfway through sampling, the form closes or you have to pay. With Porsino: Building and publishing are free, and the free account gets 50 responses a month; more responses unlock with any monthly, quarterly or yearly plan, and no response is ever lost. To make up for attrition, collect a few dozen questionnaires beyond your sample size.

Every form gets a short link made only of digits and hyphens (the Jalali creation date plus a number), so it never turns into a long jumble of characters in SMS or messaging apps. Send that link on Telegram, Eitaa, Bale or class groups; for Telegram and WhatsApp there's a direct share button too.

For a class or in-person session, put the form's QR on a slide or a handout, and if you have a website or blog, embed the form in the page with an iframe. Respondents fill in the form without creating an account. If the questionnaire is long, try the page-by-page or conversational layout; it gets a better completion rate on mobile.

Turn on “one response per person” in the publish settings. The system uses cookies and browser fingerprinting to detect that this device has already responded. That stops ordinary repeats, but someone who switches browsers or devices may respond again; that's why stricter options exist too:

• SMS mobile verification: each number gets exactly one completed response.
• Single-use invitations for a defined list, such as an organization's staff: each link accepts exactly one response, and you know who hasn't answered yet.
• The “Suspicious responses” report: it separates out responses under 10 seconds and matrices where every row has the same column ticked, so you can clean up before analysis.
• Default anti-spam — a honeypot field and a minimum fill time — filters out bot submissions.

From the “Export” menu in the responses section, download an SPSS (.sav) file; it is ready for analysis: each variable is named after the question key, the question text becomes the variable label, and options get numeric codes with value labels. Each row of a Likert matrix is a separate ordinal variable, multiple-choice questions get a 0/1 variable for each option, and the measurement level (nominal, ordinal or scale) is already set.

SPSS has no Jalali (Persian) calendar, so every date comes as a standard SPSS date with a Jalali text column next to it. A blank answer is missing data, not zero, and “No opinion” in a matrix is a user-defined missing value with its own label. The SPSS export is included in the free account, just like Excel.

1. 1

   Export to SPSS

   In the responses section, choose SPSS from the “Export” menu; the date range you picked at the top of the page applies to the file too.

2. 2

   Open in SPSS

   Open the .sav file directly; labels and codes are ready, so there is no need to code the options by hand.

3. 3

   Recode reverse-scored items

   If your questionnaire has reverse-scored items, reverse just those with Recode; in Porsino's standard questionnaires the total score is already computed with the reversed items and has its own column.

4. 4

   Cleaning and analysis

   Set incomplete and suspicious responses aside using the “status” column, compute the score for each dimension, and move on to reliability and hypothesis testing.

If you work in Excel, there is an Excel (xlsx) or CSV export too: each question is a column and each item of a Likert matrix has its own column; dates come in both Jalali and Gregorian, and the CSV file is saved so that Persian letters don't break in Excel. For reliability, paste the item columns into the Cronbach's alpha calculator.

Before exporting, take a look at the ready-made reports: descriptive statistics for each question, the mean of each matrix row, a cross-tab of two questions and a correlation matrix of numeric questions. They don't replace your thesis's statistical analysis, but they surface data errors and early patterns quickly.

A research questionnaire measures a construct, such as job satisfaction or anxiety, with several items, and for a thesis or paper it needs validity, reliability and statistical analysis. A survey serves an operational decision and each of its questions is read on its own: this month's customer satisfaction, feedback on an event or a choice between options. If yours is the second kind, the Build an online survey page explains invitations, NPS and satisfaction reports.

For the full procedure, from the informed-consent text to the SPSS export and ready-to-use wording for your methods chapter, read the guide to running your thesis questionnaire online. There are a few ready-made research questionnaires too: Lifestyle and daily habits research questionnaire, Household grocery shopping behavior study and Household financial habits questionnaire.