CIS - Centro de Investigaciones Sociológicas

1 Introduction

The objective of the Consumer Confidence Index (CCI) is to anticipate citizens' consumption behaviors. Therefore, its construction is based primarily on questions about consumer perceptions regarding both recent economic developments (6 months) and their potential developments in the immediate future (6 months). The CCI, conducted by the CIS (National Commission for the Promotion of Consumer Confidence)—and which until November 2011 was carried out by the Official Credit Institute (ICO)—has its main methodological basis in those developed by the University of Michigan and the Conference Board in the United States.

2 Survey

From September 2004 to March 2012, the CCI was developed based on a monthly opinion survey conducted by telephone using a standardized questionnaire to a representative sample of 1,000 individuals aged 16 and over residing in Spain. Starting in April 2012, the CIS decided to partially modify the sample development criteria to adapt them to the rest of the studies it conducts. The sample size increased to 1,200 interviews with individuals aged 16 and over, continuing with two-stage and stratified sampling.

Despite these changes, the sample presented a growing problem of coverage loss, as all interviews and household and individual selection were conducted via landline calls. With this sampling approach, in 2012, the study reached 81.7% of the population, leaving out the 17.8% of residents with mobile phones who did not have a landline at home. Therefore, starting in May 2013, the ICC sample was increased to 1,400 interviews, distributed between 230 interviews with individuals with mobile phones but no landline at home, and 1,170 interviews with individuals with landlines at home. Thus, coverage in terms of ICT-Households at that time stood at 99.5%, compared to the previous 81.7%.

However, the continued growth in the number of households without a landline requires a periodic review of the sample design adopted in 2013. Therefore, in 2014, the CIS decided to set a sample size of 1,510 interviews, with 1,200 interviews for households with a landline and 310 for households without a landline. Subsequently, in 2018, the total sample size was increased to 2,200 interviews. Thus, as of September 2018, the sample consists of 2,200 interviews: 1,752 for households with a landline and 448 for households without a landline. In September 2020, the sample size was increased to 3,000 interviews, initially distributing 2,287 interviews to households with a landline and 713 to households with only a mobile phone. This distribution has been modified according to the new population data available, and the latest change was made in November 2024, establishing the distribution at 1,638 interviews with households that have a landline telephone and 1,362 interviews with households that only have a mobile telephone.

The sampling procedure differs slightly depending on the type of telephone used. For landlines, stratified sampling is used. In each stratum, the household telephone number is randomly selected from the available telephone lists, and individuals are then selected according to cross-quotas for sex and age, as well as activity (employed, unemployed, or inactive). For mobile phones, random number dialing (RDD) is used, applying the same quotas as for landlines to the selection of individuals.

The strata are formed by crossing the 17 autonomous communities, along with the autonomous cities of Ceuta and Melilla, with the habitat size of the municipalities divided into 7 categories: less than or equal to 2,000 inhabitants; from 2,001 to 10,000; from 10,001 to 50,000; from 50,001 to 100,000; from 100,001 to 400,000; from 400,001 to 1,000,000; and more than 1,000,000 inhabitants. The margin of error for the total sample and under the assumption of simple random sampling is +1.8%, for a confidence interval of 95% and assuming maximum uncertainty (p=q=0.5).

Telephone interviews are preferably conducted Monday through Saturday, from 9:30 a.m. to 9:30 p.m., avoiding Sundays and holidays. Fieldwork begins in the second half of each month.

The structure of the questionnaire applied can be divided into four groups of questions:

• Questions used to calculate the ICC (questions: 3, 7, 8, 9, 12 and 14)
• Questions seeking explanations of the answers to the questions used for calculating ICC (questions: 4a, 4b, 13a, and 13b)
• Questions that can help interpret the resulting ICC (questions: 1, 2, 5, 6, 10, 11, 15, 16, 17 and 18)
• Sociodemographic questions: age, sex, education, occupation, household structure, income, ideological position, vote, etc.

3 Calculation of ICC and its components

The ICC is composed of two partial indices:

• The Current Situation Index refers to consumers' assessment of the current economic situation, relative to its situation six months earlier. It is calculated from the distribution of responses to the following questions:
  • Do you consider that the current economic situation of your family/household is better or worse than six months ago?
  • Do you think the situation in Spain for finding or improving a job is better or worse than it was six months ago?
  • Do you think the current situation of the Spanish economy is better or worse than it was six months ago?
• The Expectations Index reflects consumers' expected economic situation six months from now. It is calculated from the distribution of responses to the following questions:
  • Do you think the situation in Spain for finding or improving a job in six months will be better or worse than it is now?
  • Do you think your household situation in six months will be better or worse than it is now?
  • Do you think the Spanish economy will be better or worse in six months than it is now?

Each partial index (current situation and expectations) is the result of the following steps:

1. For each of the questions indicated, the difference between the percentage of responses indicating improvement and the percentage indicating deterioration is obtained. The other possible responses ("same," "don't know," and "no answer") do not count toward the calculation. Therefore: percentage of "better" responses minus percentage of "worse" responses.
2. Add 100 to the result obtained in the first step. This is how the result for each question is calculated.
3. Each partial index (current situation or expectations) is calculated as the arithmetic mean of the results of the three questions that comprise it. Thus, the current situation partial index is configured as follows:
   1. % “better” - % “worse” to the question: Do you consider that the economic situation of your family/household is better or worse than six months ago?
   2. % “better” - % “worse” to the question: Do you think the situation in Spain for finding/improving a job is better or worse than six months ago?
   3. % “better” - % “worse” to the question: Do you consider the current situation of the Spanish economy to be better or worse than six months ago?
   Current Situation Index = [(A+100)+(B+100)+(C+100)] / 3

Finally, to obtain the ICC, the arithmetic mean of the two partial indices (current situation and expectations) is calculated:

Therefore, the CCI value ranges from 0, when all respondents answered "worst" to the six questions considered, to 200, when all respondents selected "best" to those six questions. Taking this range into account, the CCI sets its equilibrium point at 100; values below this figure indicate an unfavorable perception of economic activity, and values above 100 indicate a favorable perception. However, beyond these values, the CCI aims to anticipate increases or decreases in household consumption, measured with a longer lag by national accounting and other statistical sources. These variations also depend on the reference point adopted.