AARP's national survey on multi-level marketing participation

Published: 30 October 2019| Version 1 | DOI: 10.17632/h3xzbzmf39.1


A 69-question survey instrument was developed by the study authors. Questions were selected based on the beliefs, values, and experiences shared by MLM participants in four focus groups conducted in an earlier phase of the research, and also on previous literature describing the characteristics of individuals and communities exposed to MLM. The survey was administered to participants belonging to the GfK KnowledgePanel, the largest online panel that uses probability-based sampling techniques for recruiting a nationally-representative sample of Americans. This sampling frame allows researchers to produce statistically valid estimates that are generalizable to the U.S. population. The survey consisted of two phases: 1) initial screening of 7,949 KnowledgePanel members to calculate lifetime prevalence of MLM in the United States; and 2) the main survey with study-eligible respondents aged 18 years and older. Qualifying respondents who participated in phase 2 included 601 KnowledgePanel members who were distributors for an MLM firm sometime in the past (adopters), and 415 KnowledgePanel members who never participated in MLM before (non-adopters). Sample size quotas of 600 MLM participants and 400 non-participants were set a priori, although more panel members were screened into the survey to accommodate for case deletions due to response error and missing data. For the 601 respondents who previously participated or currently work as distributors with an MLM firm, the survey instrument asked detailed questions about the first MLM firm they ever joined. They were asked to identify the name of the MLM company from a dropdown list (or free entry), who recruited them, the reasons why they joined, the average number of hours they worked each week, how long they maintained their membership, how much money they spent on inventory, training, and marketing materials, whether they made a profit, broke even, or lost money, and whether they felt the company accurately represented their chances of achieving financial success. The survey also gathered information on these adopters’ life circumstances, household income, age and employment prior to joining, and their reasons for leaving the MLM company if they were no longer active. They were also asked how many MLM companies they have joined in their lifetime and if they were currently working as a distributor for an MLM firm. The 415 respondents who said they have never joined an MLM firm were asked if anyone had ever asked them to join, and if so, to select one or more reasons for why they declined. Both groups were asked demographic, economic, mindset, and social activity questions.


Steps to reproduce

*recode demographics; *recode "refused" to missing and -9 to missing; array a[54] A01_1 -- C01_9 D01_1 -- D01_6 E02 E06 E08a E12 E20 E26 E11 f01 f02 f03 f04 f07 G01 G03 E09a E13a E13b E14a E14b E14c E16 E17 E19 E21; do i= 1 to 54; if a[i] = -1 then a[i]=.; if a[i] = -9 then a[i]=.; if a[i] = -8 then a[i]=.; end; drop i; *recode "refused" to missing; array b[1] E09b; do i= 1 to 1; if b[i] = 6 then b[i]=.; end; drop i; *recode "no" to 0, "yes" to 1; array c[8] E01 E08b E15 E18 E22 G02 E24 E25; do i= 1 to 8; if c[i] = 2 then c[i]=0; if c[i] = 1 then c[i]=1; if c[i] = -1 then c[i]=.; end; drop i; array d[1] C01_4; do i=1 to 1; d[i]=6-d[i]; end; MLMever=.; label MLMever= ever participated in MLM; if DOV_MLM=1 then MLMever=1; if DOV_MLM=2 then MLMever=0; approached=E26; label approached= ever asked to participate in MLM (solicited); if MLMever=1 then approached=2; NumberMLM=E02; label NumberMLM= number of MLMs ever joined; if MLMever=0 then NumberMLM=0; TwoplusMLM=E02; if E02 ge 2 then TwoplusMLM=1; else if E02 = 1 then TwoplusMLM=0; ThreeplusMLM=E02; if E02 ge 3 then ThreeplusMLM=1; else if E02 le 2 then ThreeplusMLM=0; CurrentMLM=E01; label CurrentMLM= currently working in MLM (any); if MLMever=0 then currentMLM=0; nine5=5-A01_6; label nine5= dont want a '9 to 5' job; material=6-mean(C01_1, C01_2, C01_3, C01_4, C01_5, C01_6, C01_7, C01_8, C01_9); label material= mean materialism values score; mindset=5-mean(A01_1, A01_2, A01_3, A01_4, A01_5, nine5); label mindset= mean of work attitudes w/o risk taking item or '9-5' item; reward=5-mean(B01_1, B01_2, B01_3, B01_4, B01_5); label reward= mean of reward responsiveness; drive=5-mean(B01_6, B01_7, B01_8, B01_9); label drive= mean of drive; pitches=6-mean(D01_1, D01_2, D01_3, D01_4, D01_5, D01_6); label pitches= mean interest in MLM recruitment pitches; lifevent=sum(E10_1, E10_2, E10_3, E10_4, E10_5, E10_6, E10_7, E10_8, E10_9, E10_10, E10_11, E10_12, E10_13, E10_14); label lifevent= sum of life events; negevent=sum(E10_1, E10_2, E10_5, E10_6, E10_7, E10_10, E10_11, E10_13); label negevent= sum of negative life events; services=6-f02; label services=frequency of religious participation; communit=6-f03; label communit=frequency of community engagement; media=9-f04; label media=frequency of social media use; else if F07 gt 1 then donald=0; bankrupt_after=.; label bankrupt_after=declared bankruptcy during or after MLM; if G03=2 | G03=3 then bankrupt_after = 1; else if G03=1 then bankrupt_after = 0; else if G02=0 then bankrupt_after = 0; fin_fragility=.; label fin_fragility=probably or definitely could not cover a one-thousand dollar emergency expense; if G01=1 | G01=2 then fin_fragility = 0; else if G01=3 | G01=4 then fin_fragility = 1; run;


Social Sciences, Marketing, Consumer, Cognitive Bias, Sales, Fraud