Sign in
x
☰

Extract common text from titles

Extract repetitive text in a series of similar questions

Updated at 2019-06-26 16:40:01 UTC

Many surveys have questions, like "For each of the following..." or "Check all that apply...". When you export to SPSS survey engines combine the text for each individual question with the text instructions for the question group. 

In q5 there are a series of questions q5a, q5b, q5c, and q5d. Each share the common title text "In your opinion, in AMERICA, how much conflict is there between...". Protobi has a feature to get rid of repetitive text by removing it from the child elements and putting it at the group level.

1. Select the parent group by pressing its header.

Protobi interface displaying a group of four related survey questions (q5a through q5d) under parent element q5. Each sub-question begins with identical text "In your opinion, in AMERICA, how much conflict is there between..." followed by different demographic comparisons. Each shows horizontal bar charts with percentage distributions across conflict level responses.

2. From the Advanced menu choose "Extract common text" or press SHIFT + E + X on your keyboard.

Protobi interface showing the Advanced menu dropdown opened with "Extract common title text" option highlighted. The menu appears over the data view showing the same four survey questions. The menu displays various editing options including Group, Dissolve, Clone, Change key, Extract common title text, Autogroup, and other element management functions.

 

3. Choose the text you want to extract and press ok.

Protobi finds the longest strings that appear in all of the questions, and presents these as suggested options. You can also manually enter a specific string in the input box.   

Dialog box titled "Extract common text" showing three radio button options for text to extract. The first option "In your opinion, in AMERICA, how much conflict is there between... -" is unselected, the second longer option is selected (shown with filled radio button), and the third option allows custom text entry via an input field labeled "text fragment". Cancel and Ok buttons appear at the bottom.

4.Protobi will bring the text to the title of the parent group and remove the matching string from each child if found.

Protobi data view after extraction showing parent group q5 now displays the extracted common text in its title. The four sub-questions (q5a-q5d) now show only their unique portions: "a Young people and older people", "b Poor people and rich people", "c Blacks and whites", and "d Immigrants and people born in the United States". Each retains its bar chart with percentage distributions.

See Extract common text from titles for an animated GIF.  

With the common text removed, the short labels appear nicely when in compact display (please see "Compact to..." tutorial):

Compact view of question group q5 showing the extracted common text in the parent header. The four sub-questions now appear as a condensed list with short labels ("a Young people and older people", "b Poor people and rich people", "c Blacks and whites", "d Immigrants and people born in the United States") each with a single percentage value and blue bar displayed horizontally.