Help Center

See Crunch Automation basics for more information.

The CREATE MULTIPLE DICHOTOMY FROM CONDITIONS command allows you to create a multiple response variable indicating the logical condition and optionally indicate the base for the evaluation, either by indicating the VALID rows or MISSING rows as a logical expression.

It is possible to use a global VALID/MISSING expression for all conditions or enter them individually, but both cannot be used simultaneously.

Each response statement must contain a LABEL and an optional ALIAS:

CREATE MULTIPLE DICHOTOMY FROM CONDITIONS
    MAPPING
        SELECT condition [(MISSING|VALID) condition] LABEL "label" [ALIAS alias],
        ....
        SELECT condition [MISSING|VALID) condition] LABEL "label" [ALIAS alias]
    END
    [MISSING|VALID) condition]
AS alias
[TITLE "string"]
[DESCRIPTION "string"]
[NOTES "string"];

Example Syntax

CREATE MULTIPLE DICHOTOMY FROM CONDITIONS
    MAPPING
        SELECT A=1 VALID is_valid(A) LABEL "First" ALIAS s1,
        SELECT B=1 OR C = 1 MISSING B != 1 LABEL "Second",
        SELECT D != 1 AND (C != 1) LABEL "Third" ALIAS s2
    END
AS my_mr_cond;

CREATE MULTIPLE DICHOTOMY FROM CONDITIONS
    MAPPING
        SELECT A=1 LABEL "First" ALIAS s1,
        SELECT B=1 OR C = 1 LABEL "Second",
        SELECT D != 1 AND (C != 1) LABEL "Third" ALIAS s2
    END
    MISSING is_missing(asked) OR asked = 1
AS my_mr_cond;

Use Case - Brand Funnel

Survey with a Single Brand:

For the example that follows, the survey contained the following questions for a particular brand:

• Q8 - "Prior to taking this survey, had you heard of BRAND?" (single response - 1=Yes, 2=No)
• Q9 - "You indicated that you have heard of BRAND, how familiar would you say that you are with BRAND?" (single response - 1=Very familiar, 2=Somewhat familiar, 3=Not at all familiar)
• Q10 - "Have you seen or heard advertisements or information about BRAND in the past 3 months?" (single response - 1=Yes, 2=No)
• Q11 - "Have you ever purchased any of BRAND's products?" (single response - 1=Yes, 2=No, 3=Don't know)
• Q12 - "How likely would you be to purchase any BRAND products again?" (single response - 1=Very likely, 2=Somewhat likely, 3=Not at all likely)

In the following example, you are building a brand funnel for the brand. For the new variable, some of the categories should be based on how the respondent answered previous questions, which may not have been included in the survey logic, so you define a VALID criteria for those categories. For familiarity and likely to purchase, you want to include those who answered very or somewhat, so you include the IN [1,2] syntax.

CREATE MULTIPLE DICHOTOMY FROM CONDITIONS
  MAPPING
    SELECT Q8=1 LABEL "Awareness" ALIAS brand_funnel_1,
    SELECT Q9 IN [1,2] VALID Q8=1 LABEL "Familiarity" ALIAS brand_funnel_2,
    SELECT Q10=1 VALID Q8=1 LABEL "Seen Advertisements" ALIAS brand_funnel_3,
    SELECT Q11=1 VALID Q8=1 LABEL "Purchased" ALIAS brand_funnel_4,
    SELECT Q12 IN [1,2] VALID (Q8=1 AND Q11=1) LABEL "Repeat Purchase" ALIAS brand_funnel_5
  END
AS brand_funnel
TITLE "Brand Funnel";

Survey with Multiple Brands:

For the example that follows, the survey contained the following questions with a list of brands as the answer options:

• Q4 - "Which of the following brands have you heard of before?" (multiple response)
• Q5 - "From which of the following brands would you consider purchasing items?" (multiple response)
• Q6 - "Which brand do you prefer?" (single response)
• Q7 - "Which of these brands, if any, have you purchased in the past three months?" (multiple response)

Since Q4, Q5, and Q7 are multiple response questions, when you need to refer to a single element ("subvariable") of an array alone (not explicitly in the context of the array it is a member of), you will now need to use the array_alias[subvariable_alias] syntax to provide the necessary indication of where the subvariable can be found (i.e., inside which array). When doing so, it requires that you check the "Strict subvariable syntax" checkbox at the bottom of the Crunch Automation panel.

Set the Base of Each Category to Those Aware of the Brand 

In the following example, you are building a brand funnel for the second brand in the list. For the new variable, perhaps you want each category to be based to those who were aware of that brand in Q4, so you define a VALID criteria for the last 3 categories.

CREATE MULTIPLE DICHOTOMY FROM CONDITIONS
  MAPPING
    SELECT Q4[Q4r2]=1 LABEL "Awareness" ALIAS funnel_brandB_1,
    SELECT Q5[Q5r2]=1 VALID Q4[Q4r2]=1 LABEL "Consideration" ALIAS funnel_brandB_2,
    SELECT Q6b=2 VALID Q4[Q4r2]=1 LABEL "Preference" ALIAS funnel_brandB_3,
    SELECT Q7[Q7r2]=1 VALID Q4[Q4r2]=1 LABEL "Purchase" ALIAS funnel_brandB_4
  END
AS funnel_brandB
TITLE "Brand Funnel - Brand B";

Set the Base to the Total Sample

In the following example, you are building a brand funnel for the second brand in the list. For the new variable, perhaps you want to base each category on the total sample. In this case, you do not need to define a VALID criteria for each category, you can assign it to the entire variable. To get the total base, you can refer to a variable where each respondent had a value, in this case Q1, but it could be any variable you'd like to use.

CREATE MULTIPLE DICHOTOMY FROM CONDITIONS
  MAPPING
    SELECT Q4[Q4r2]=1 LABEL "Awareness" ALIAS funnel_brandB_1,
    SELECT Q5[Q5r2]=1 LABEL "Consideration" ALIAS funnel_brandB_2,
    SELECT Q6b=2 LABEL "Preference" ALIAS funnel_brandB_3,
    SELECT Q7[Q7r2]=1 LABEL "Purchase" ALIAS funnel_brandB_4
  END
  VALID is_valid(Q1)
AS funnel_brandB
TITLE "Brand Funnel - Brand B";