This page presents additional analysis of survey data collected by the Mongolian Red Cross (MRCS) covering questions around trust as part of the Community Trust Index project. The data was collected in summer 2023, covering 12 regions and 125 districts of the country.

Sampling

Coverage

The sampling approach used was a non-probability sampling, that at least in part used contact information of an existing database of people associated and acquainted with the Mongolia Red Cross Society. The sample is skewed towards the rural population, with 58% of the sample being classified as rural, while according to World Bank data that rural population is closer to 30%.

It is important to note that this non-probability design does not allow for inferences to be made about the general population in Mongolia. Therefore, all the data should be regarded as indicative, and it is strongly advised against presenting this data as representative of the entire country of Mongolia, as the convenience sampling approach taken does not allow for drawing generalizations for the population in Mongolia.

However, a post-stratification technique can be employed to ensure that the weighted sample corresponds to some features of the population more closely. Therefore, a initial analysis focuses on a few demographic parameters of our sample and compares them to the overall population in the country.

Gender and Age

Compared to the overall Mongolian population, in the sample women are over-represented, while younger people, especially men are under-represented.

Education

When it comes to education, people with higher levels of education are over-represented in the sample, in particular with university degrees and those who graduated form secondary school.

Employment

While it was difficult to source matching data on work employment, available data sourced from the World Bank suggests that people with employment in agriculture are underrepresented in the sample, which matches the larger share of people with higher education in the sample.

NS relationship

Since the sample is skewed towards those who have a history/relationship with the Mongolian Red Cross Society, it is to be expected that a fairly large share of the sample has donated to the Mongolian Red Cross Society, requested and received support. We indeed see that in our sample with 40% of the people having volunteered or made a donation, and more than 25% having received support.

Results

Distribution of mean scores for overall competency and values per demographic questions

When looking at the overall results for competency and values, we see no major differences between age groups, gender, and rural/urban settings. The largest differences we see among district, retired vs not retired and beneficiaries or volunteers.

Score distribution

Weighting

To at least partly address the deviation of demographic parameters from the overall population, we have utilized a technique called raking. The raking process adjusts the results based on several variables to ensure that our sample reflects the distribution of these variables in the overall population. Here are the variables we considered for raking:

  • Age groups and gender: We used UN data to determine the proportions of different age groups and gender in the population.
  • Education: We used World Bank data on education of the general population in Mongolia

Additionally, we can make assumptions for the upper bounds of volunteer and aid recipient percentages since no specific data was available. We estimated these rates to be 10% of the population to just see the impact of the large share of volunteers and beneficiaries in the sample. These rates of 10% are just estimates and we will not use this raking scenario for further analysis. It can, however, help to to understand better the impact of the biased sample.

Using an appropriate package in R to conduct the raking, we obtained the following results:

  • Raking scenario 1: Raked by age groups, gender, education.
  • Raking scenario 2: Scenario 1 with additional raking by the percentage of respondents who volunteered (assuming 10% of the population volunteered).

The results indicate that the un-weighted results provide the most positive ratings across the questions. If we cap the percentage of volunteers and beneficiaries, results are lower compared to raking using just general demographic values (scenario 1).

Weighted Results

The weighted results show little variation among the competency questions that range from 2.2-2.4 on a 0-3 scale. Questions on values show more variation ranging from 1.9 for transparency to 2.5 on humanism. Transparency, neutrality as well as engagement are the questions with the lowest perceptions and might be areas that the MRCS could focus on to improve composite perceptions in the value dimension realm of the trust index.

Score per profile

When looking at the sub-groups of people who volunteered and beneficiaries as well as others, we see that people who have received support from the MRCS provide the most positive rating from the three groups. For competency questions the results show no variation in the sense that the non-volunteer/beneficiary group always provides the lowest rating, beneficiaries the highest, with a small gap to volunteers, which however is not significant. For the value question we see a similar pattern, again differences between beneficiaries and volunteers are not significant.

Download Summary Excel File

Significance testing

When checking for significant differences between the groups we use a t-test to compare means of the competency and value questions, we see that for volunteers and beneficiaries the results are not significantly different. In general, the table shows whether a results for beneficiaries, volunteers and others are significantly different form each other. We used a 99% confidence level and corrected the p-values using a multiple comparisons correction.

Dimension Drivers Volunteer-Other Volunteer-Beneficiary Benficiary-Other
Competency Capabillity Yes Yes Yes
Competency Responsiveness Yes No Yes
Competency Awareness Yes No Yes
Competency Accessible Yes No Yes
Competency Openess Yes No Yes
Competency Relevance Yes No Yes
Competency Effectiveness Yes No Yes
Value Kindness Yes No Yes
Value Fairness Yes No Yes
Value Inclusiveness Yes No Yes
Value Engagement Yes No Yes
Value Integrity Yes No Yes
Value Transparency Yes No Yes
Value Neutrality Yes No Yes
Value Humanism Yes No Yes