Conclusion strategies and bad group decision-making: a group coming together experiment

Kazuhisa Takemura , in Escaping from Bad Decisions, 2021

Assay of overall satisfaction for discussion

Multiple regression analysis was conducted to examine the bear on of the three factors of decision-making strategy, the grouping to which the participants belonged to, and the type of agenda on overall give-and-take satisfaction. As a result of comparing and ranking the AIC of each model, the model with the lowest AIC predicted the satisfaction of the entire word by the interaction of the conclusion strategy and the grouping to which the participant belonged to (model number 9), with an AIC of 3096.21. In other words, among the 11 models examined, the model that predicts the satisfaction of the entire discussion by the interaction of the determination strategy, the group to which the participant belongs to, and the group to which the participant belongs to with the decision strategy can exist judged to be the model with the highest prediction ability.

The fractional regression coefficients of the determination strategies in this model are shown in Table 11.4, where the partial regression coefficients for DIS were −0.11 for WAD and −0.04 for LEX [LEX: t(1140)=−0.31, due north.s.]. This suggested that the overall satisfaction with the discussion did non change when comparing WAD to DIS and LEX to DIS, respectively.

Table eleven.four. Akaike information criterion rank and partial regression coefficient o for decision strategy, grouping, and agenda.

Item Model number Model DIS WAD LEX
Procedure evaluation ix Intercept+decision strategy+group+decision strategy × group i.00 0.07 0.02
Result evaluation 9 Intercept+decision strategy+group+decision strategy × group ane.00 0.19 0.06
Satisfaction 9 Intercept+decision strategy+group+conclusion strategy × group 1.00 0.11 0.04

Notes: Grouping: group in which participants belonged to, decision strategy × grouping: interaction between decision strategy and group, process evaluation: evaluation of the give-and-take process, outcome evaluation: evaluation of outcome, and satisfaction: satisfaction level of discussion. DIS, Disjunctive; LEX, lexicographic; WAD, weighted-additive conclusion.

Read full affiliate

URL:

https://www.sciencedirect.com/science/article/pii/B9780128160329000090

Results, Word, and Conclusion

Katerina Petchko , in How to Write Almost Economic science and Public Policy, 2018

Results of Multiple Regression Analysis (MRA)

Recall that MRA is a statistical process that assesses the relationship between a dependent variable and several predictor variables. The estimates generated by MRA are called coefficients. Using MRA, we can calculate the corporeality of variance in the dependent variable that is accounted for (= explained) past the variation in each of the independent variables. This calculation shows the relative importance of each independent variable to the relationship.

It is across the telescopic of this book to provide a detailed treatment of MRA as a statistical technique. For a basic estimation of MRA results in economics, consult Greenlaw (2009). For advanced data on MRA and other statistical techniques, you may wish to consult Tabachnick and Fidell's Using Multivariate Statistics.

In an MRA study, the post-obit information generated past regression software is normally reported.

•

The size and sign of regression coefficients. The size of regression coefficients shows how much each predictor variable contributes on its own to the variance in the dependent variable afterward the effects of all the other predictor variables in the model have been statistically removed. In their standardized course (equally β), regression coefficients are a measure out of the importance of each variable, assuasive researchers to compare the relative importance of the predictors. In economics and public policy, the sign of regression coefficients is too important and it is discussed in comparison with the expected (or hypothesized) sign predicted from theory: Do the explanatory variables have the expected sign?

•

Statistical significance for each estimated coefficient, which is determined by comparing the p-value (or significance probability) associated with a coefficient with the chosen level of significance. If the p-value is smaller, the coefficient is interpreted every bit being statistically significant; if it is greater, the coefficient is interpreted as being nonsignificant, or every bit not beingness significant. There are many variations in the reporting and estimation of nothing hypothesis significance testing in public policy and economics. For example, in economics, three significance levels are commonly used: ane%, 5%, and ten% and results are frequently described as being "statistically pregnant at the one% (or 5%, or x%) significance level." The ten% significance level is uncommon in other disciplines, for example, in sociology or education, where results with p-values that are greater than .05 (five%) are interpreted as being nonsignificant.

Alternatively, when reporting statistical significance, researchers may simply indicate whether the generated p-values are smaller than the level of significance. In this example, authors signal statistically significant values with asterisks—a single asterisk (⁎) for p  <   .01, a double asterisk (⁎⁎) for p  <   .05, and a triple asterisk (⁎⁎⁎) for p  <   .1—and use a note nether the table to evidence what the asterisks refer to.

In some enquiry areas, authors may provide the exact p-values (e.g., p  =   .58). Providing the exact p-values is especially mutual in psychological and educational research, but it is fairly uncommon in economics. In some areas, confidence intervals are normally used to point significance levels.

Because of the great variability among disciplines in reporting statistical significance, it is important to detect out what is common in the particular area you are working in and report statistical results using the conventions of your field.

•

"Goodness-of-fit" statistics. These statistics show how well the model you are testing explains the information: How much variance in the dependent variable is explained past the combination of the predictors? The F-statistic is used to determine if all the coefficients in the model are statistically significant, whereas R two (or adapted R two) is used to decide the overall corporeality of variance in the dependent variable that is explained by all the predictor variables in combination.

Greenlaw (2009, p. 217) gives good advice for interpreting R 2: "R 2 for cross-section data is by and large less than R two for time-series data. Econometricians typically consider a fourth dimension-series regression to be "practiced' if it results in an R ii of 0.8 or college. By contrast, a cantankerous-section regression is considered "good" if information technology has an R 2 of only half that: 0.4 or higher up."

Regression results are always presented in table form. A typical regression tabular array includes the post-obit information: regression coefficients, standard errors (in parentheses), statistics indicating significance, and goodness-of-fit statistics. Information technology is important to stress here that regression tables that are included in a paper are always constructed and never copied straight from regression output provided by the regression software. Later in this section, I give suggestions for formatting tables in a quantitative study.

Read full chapter

URL:

https://www.sciencedirect.com/scientific discipline/article/pii/B9780128130100000144

Projections and Risk Assessment

Morton Glantz , Johnathan Mun , in Credit Engineering for Bankers (Second Edition), 2011

xi Multiple Regression

To run the multiple regression assay, follow these steps:

1.

First Excel and open the example model Risk Simulator | Example Models | 01 Advanced Forecast Models.

2.

Go to the Regression worksheet.

3.

Select the data area including the headers or cells B5:G55 and click on Risk Simulator | Forecasting | Regression Analysis. Select the Dependent Variable as the variable Y, leave everything else alone, and click OK. Review the generated study.

•

Exercise Question: Which of the independent variables are statistically insignificant, and how can you tell? That is, which statistic did you lot use?

•

Practice Question: How proficient is the initial model'southward fit?

•

Exercise Question: Delete the entire variable columns of data that are insignificant and rerun the regression (i.e., select the column headers in Excel'due south grid, right-click, and delete). Compare the R-Square and Adjusted R-Foursquare values for both regressions. What can you determine?

•

Exercise Question: Will R-square always increase when you have more than independent variables, regardless of their being statistically significant? How about adjusted R-square? Which is a more than bourgeois and appropriate goodness-of-fit measure?

•

Practice Question: What can yous practice to increase the adjusted R-Square of this model? Hint: Consider nonlinearity and some other econometric modeling techniques.

•

Exercise Question: Run an Auto-Econometric model on this dataset and select the nonlinear and interacting selection and see what happens. Does the generated model ameliorate fit the data?

Read total chapter

URL:

https://world wide web.sciencedirect.com/science/article/pii/B9780123785855100089

Research Proposals

Katerina Petchko , in How to Write About Economic science and Public Policy, 2018

Methodology

A quantitative paper states a hypothesis and tests it using statistical tools (eastward.g., multiple regression analysis) to produce generalizable results. If yous are writing a quantitative proposal, explain what kind of data you will use and where the data will come from. Explicate your empirical methodology, also what model y'all will utilize, what variables you volition include, and how yous will measure them.

A qualitative empirical paper explores a phenomenon or a process using multiple sources of information including in-depth interviews, documents, and observation. If yous are writing a proposal for a qualitative written report, explain why and/or how you selected your case(s), what data you programme to utilize, and how you volition collect the information.

Read full chapter

URL:

https://www.sciencedirect.com/science/commodity/pii/B9780128130100000077

Research in Public Policy and Economics

Katerina Petchko , in How to Write About Economic science and Public Policy, 2018

Which Approach Is Prevalent in Public Policy Programs?

Since the 1970s, public policy literature has been characterized by the pervasive use of quantitative methods of data collection and analysis including survey enquiry, quasiexperimental enquiry, multiple regression assay, cost-benefit analysis, and economic modeling. Although public policy research became more than diversified in the 1990s ( Radin, 2000) and began to include qualitative studies, quantitative enquiry remains prevalent in public policy and, peculiarly, policy analysis, both in journal publications and in educational curricula. For instance, in a review of educational curricula of 44 programs in public policy and policy analysis taught at leading public policy schools in the United States, Morçöl and Ivanova (2010) found that quantitative courses constituted an overwhelming majority of courses taught at both the master's (88%) and doctoral (79%) levels. They also found that the most frequently taught method of information collection was survey and the most oftentimes taught method of data assay was multiple regression assay. This emphasis on quantitative methods is as well reflected in the predominantly quantitative types of papers that students in public policy programs are oft required to write.

Read full chapter

URL:

https://www.sciencedirect.com/science/commodity/pii/B9780128130100000028

Data and Methodology

Katerina Petchko , in How to Write About Economics and Public Policy, 2018

Quantitative vs. Qualitative Data Assay

There are many dissimilar strategies and techniques for data analysis in public policy and economics; some are more than common in a particular enquiry expanse than others. The choice of the particular strategy for data analysis is dictated largely by the purpose of your study—by its research question—and past the type of data that are used.

Quantitative research questions usually ask about relationships among multiple variables, and data are commonly observational rather than experimental. Past far, the most common tool used to analyze such data is multiple regression analysis . Multiple regression analysis allows researchers to assess the strength of the relationship between an outcome (the dependent variable) and several predictor variables as well as the importance of each of the predictors to the human relationship, often with the effect of other predictors statistically eliminated.

It is important to betoken out, still, that multiple regression assay is a statistical technique, non a inquiry pattern, and as such, it does not establish causation. This is because multiple regression builds on correlation, which shows mere associations between variables. To infer a causal relationship, re- searchers demand to eliminate bias resulting, for case, from variables that cannot be observed. This can be washed by design—through experimental manipulation of variables, or by using statistical controls. The second pick is much more than common in studies of public policy and economics. Diverse approaches can be used to minimize bias due to reverse causality and omitted variables. Console regression with fixed furnishings is one example of a commonly used approach in economics enquiry. All the same, console regression requires the employ of panel information, which may not always be available, and they, too, have limitations. It is, therefore, wise to proceed in mind when interpreting results, that fifty-fifty under the best of circumstances, statistical controls are never fool-proof.

Qualitative data analysis in public policy depends on whether the written report is information-based or literature-based. In data-based studies (eastward.g., studies based on information collection through interviews, focus group discussions, or participant observation), data assay involves transcribing and coding participants' responses and/or the researcher's notes by identifying certain themes or patterns in the data that help answer the research question(s). In many means, qualitative data analysis is an attempt to reduce a very large amount of qualitative information—participants' responses and comments—to a few themes. For example, if your written report has looked at how poor women in rural areas cope with violence, you may want to analyze the women's responses to identify the strategies that they have used. You would have to make many subjective decisions about what the women's responses really mean and you would demand to be very clear nigh how you made those decisions. Using multiple sources of data (e.one thousand., interviews   +   documents   +   observation) in a qualitative study is one strategy to reduce subjectivity.

In literature-based studies, there are usually no data and a paper may exist based solely on summarizing an often capricious selection of studies or other documents. In such studies, information technology is mutual for authors to explain how the literature was located, how the specific studies and documents were selected, and how they help reply the inquiry question(s).

Read total chapter

URL:

https://www.sciencedirect.com/science/article/pii/B9780128130100000132

Hypothesis Testing

Laura Lee Johnson , ... Pamela A. Shaw , in Principles and Practice of Clinical Research (Fourth Edition), 2018

Multiple Comparisons

When making many statistical comparisons, i.eastward., performing multiple hypothesis tests, a certain fraction of the test statistics will be statistically meaning fifty-fifty when the zippo hypothesis is true. In general, when a series of tests is performed at the α significance level, approximately α  ×   100% of tests will be meaning at the α level even when the null hypothesis for each examination is true. For instance, even if the null hypotheses are true for all tests, when conducting many independent hypothesis tests at the 0.05 significance level, on average (in the long term) 5 of 100 tests will exist significant by gamble alone. Problems of multiple comparisons arise in diverse situations, such as in clinical trials with multiple end points and multiple looks at the data. By doing multiple tests, you naturally increase your chances of making a blazon I error if no adjustment is made to the usual testing framework for a unmarried test statistic. Pairwise comparing among the sample means of several groups is also an expanse in which issues of multiple comparisons may be of business organization. For k groups, there are k(yard  – i)/2 pairwise comparisons, and just by chance some may reach significance. Our concluding example is with multiple regression assay in which many candidate predictor variables are tested and entered into the model. Some of these variables may result in a significant result just by chance. With an ongoing study and many interim analyses or inspections of the data, with no adjustment for performing multiple comparisons, we accept a high run a risk of rejecting the zilch hypothesis at some fourth dimension point even when the zip hypothesis is true.

In that location are diverse approaches to the multiple comparisons problem. First, consider if multiple comparisons is actually a problem. If we ask multiple questions we look multiple answers. If we ask related questions we expect related answers. Looking at the totality of the show when interpreting results is far more useful than overzealous correction for multiple comparisons (or ignoring all just the unmarried significant p-value out of l). One rather breezy approach to multiple comparisons is to cull a significance level α lower than the traditional 0.05 level (east.g., 0.01) to prevent many false-positive conclusions or to "command the simulated discovery charge per unit." The number of comparisons should be made explicit in the article. More formal approaches to control the "experiment–wise" type I error using corrections for multiple comparisons have been proposed. An example is the Bonferroni correction, in which the type I error charge per unit is taken as α/due north, where n is the number of comparisons fabricated. Another class of methods has been developed to right for multiple comparisons that result from monitoring trial results during the trial. Interim monitoring methods that control the type I error charge per unit are bachelor for various study designs and are discussed further in Chapter 27. 14 The classic reference by Hochberg and Tamhane provides a broader discussion of methodology to adjust for multiple comparisons. fifteen

Information technology is all-time to address the upshot of multiple comparisons during the design stage of a written report. One should determine how many comparisons will be fabricated and so explicitly state these comparisons. Studies should more often than not be designed to minimize the number of statistical tests at the end of the written report. Advertizing hoc solutions to the multiple comparisons problem may be washed for exploratory or epidemiologic studies. Multiple comparison adjustments should be made for the chief analyses of definitive studies (such as phase 3 confirmatory studies) to rigorously maintain the type I mistake charge per unit, i.e., the probability of falsely rejecting any null among those tested, at the chosen α level. Studies that focus on a single primary issue and data analyzed at the terminate of study avoid the issue of multiple comparisons. The topic of multiple comparisons is expanded in Chapter 27.

Read full chapter

URL:

https://world wide web.sciencedirect.com/scientific discipline/article/pii/B9780128499054000241

Revisiting the grouping controlling experiment

Kazuhisa Takemura , in Escaping from Bad Decisions, 2021

thirteen.4 Results

To examine the factors that crusade irrational meetings, we created eight videos of meeting scenes and conducted an experiment. Then, analyses were conducted to examine the furnishings of the effect of the coming together decision, the emphasis of the risk, and the compliance with the rules on the desirability of the meeting decision and process. Specifically, comparing of ways, analysis of variance, assay of covariance, and analysis of correlation were conducted.

13.4.1 Analysis of the desirability of a meeting decision

Outset, we calculated the means of the desirability of the decision in each video. In Fig. thirteen.6 the first left particular 1 indicates the low risk and risky consequence (Matsuzaka beef) condition. The other numbers of items are follows: Item ii for the loftier risk and risky outcome condition, Item 3 for the low risk and riskless result (Imported beef), Item 4 for the high risk and riskless upshot status, Item five for the rule noncompliance and risky outcome condition, Particular 6 for the rule compliance and risky outcome condition, Particular 7 for the noncompliance and riskless outcome status, and Item 8 for the rule compliance and riskless outcome status. For clarity, when the final determination result was Matsuzaka beef, the bar graph was ready to "Matsuzaka" and colored black, and when the final decision result was imported beef, the bar graph was set to "Import" and colored white.

Figure thirteen.6. Hateful and standard deviation (SD) for the desirability of decision in each condition.

Fig. thirteen.6 shows the mean and the SD for the desirability of decision regardless of the emphasis of the run a risk or the compliance with the rule; the desirability of the conclusion of the meeting tended to be college at the level where the effect of the determination was imported cattle. Next, we conducted a ii-level analysis of variance for each of the two between-subjects factors separately for risk intensity and conclusion effect and for rule compliance and decision outcome.

To examine the influence of the degree of chance and the decision of the meeting on the desirability of the decision, we conducted a two-level analysis of variance for each of the two betwixt-subjects factors of risk (high risk vs depression adventure) and determination outcome (Matsuzaka beef (risky issue) vs imported beefiness (riskless event)). Every bit a outcome, a pregnant main effect was plant for the conclusion upshot in the meeting (F(ane,59)=21.25, P<.001). Regardless of the accent of the run a risk, the final decision for imported beef was evaluated as more desirable than that for Matsuzaka beefiness.

Next, to examine the effects of dominion compliance and decision outcome on the desirability of the conclusion, nosotros conducted a two-level analysis of variance for each of the two between-subjects factors of dominion compliance (rule compliance vs dominion noncompliance) and decision outcome (Matsuzaka beef vs imported beef). Equally a upshot, a pregnant main upshot was constitute for the determination consequence in the meeting (F(1,57)=4.98, P<.05). Regardless of whether the dominion was adhered to or not, the final conclusion for imported beefiness was evaluated equally more desirable than that for Matsuzaka beef.

In the multiple regression analysis of the desirability of the decision consequence of the meeting, since it was not possible to analyze the iii factors together due to the experimental design, the determination outcome (Matsuzaka beef vs imported beefiness), run a risk (high vs low), and compliance with the rule (command vs rule compliance vs dominion noncompliance) were used as independent variables. Multiple regression analysis was conducted to predict the desirability of the determination from these independent variables. The command status for the presence or absence of rule compliance was used every bit the low take chances condition, in which there was no disharmonize over the rule in the majority vote. The results showed a positive fractional regression coefficient that was significant at the 0.1% level ( t=4.81, P<.001). In other words, it was recognized that the outcome of the meeting was more desirably evaluated when the final conclusion was made for imported cattle. There was a significant negative partial regression coefficient at the 5% level in the noncompliance condition (t=−2.07, P<.05). In other words, the status in which the determination was overturned due to noncompliance with the rules was rated as significantly less desirable than the control condition of rule compliance.

Nosotros conducted an analysis of covariance with the decision issue as the predictor variable, the desirability of the procedure in the coming together as the covariate, and the desirability of the conclusion in the meeting as the dependent variable. The results showed that the desirability of the meeting decision was significantly higher in the condition in which the decision outcome was imported beefiness than in the status in which the decision consequence was Matsuzaka beef (t=5.18, P<.001).

Next, we conducted an analysis of covariance with risk intensity as the predictor variable, desirability of the meeting process as the covariate, and desirability of the meeting decision as the dependent variable. The results showed that at that place was no effect of hazard intensity on the desirability of meeting decisions (t=1.30, n.s.).

thirteen.4.ii Analysis of the desirability of the coming together process

Fig. 13.seven shows the hateful and SD of the desirability of the meeting process for each video equally a bar graph. The numbers of items indicate the aforementioned weather every bit mentioned in the previous section.

Figure xiii.7. Mean and SD for the desirability of meeting process in each condition.

Unlike the desirability of the decision, no departure was found between the decision results of Matsuzaka and imported cattle. In improver, regardless of the final decision result, the evaluation tended to be higher in the rule compliance condition than in the rule noncompliance condition. Next, nosotros conducted a two-level analysis of variance for each of the two between-subjects factors, dividing the results into the emphasis of gamble and decision outcome, and the presence or absence of dominion compliance and decision outcome.

To examine the influence of risk accent and decision upshot on the desirability of the procedure, we conducted a two-level analysis of variance for each of the two between-subjects factors: take a chance (high vs low) and decision outcome (Matsuzaka vs imported beefiness). The master furnishings of risk (loftier vs low) and decision upshot (Matsuzaka vs imported beef) on the desirability of the process in the meeting and their interactions were not significant (F(1,59)=0.02, due north.s., F(i,59)=0.40, n.s.).

Next, to examine the influence of dominion compliance and decision outcome on the desirability of the process, we conducted a two-level analysis of variance between subjects for dominion compliance (adherence vs nonadherence) and decision consequence (Matsuzaka beef vs imported beef). A 2-level analysis of variance was conducted for each gene. As a result, no factor was found to affect the desirability of the procedure in the meeting.

Equally in the multiple regression analysis of the desirability of conclusion-making, nosotros conducted a multiple regression analysis to predict the desirability of the process in the meeting from the three factors of determination outcome, run a risk intensity, and dominion compliance. As a result, no independent variable was plant to influence the dependent variable (for all t <2.00, n.south.).

An analysis of covariance was conducted with the dominion compliance gene equally the predictor variable, desirability of coming together decisions as the covariate, and desirability of meeting procedure as the dependent variable. The results of the assay of covariance showed that there was no effect of the factors of noncompliance (t=−i.39, due north.s.) and compliance (t=0.69, n.s.) on the desirability of the meeting procedure.

13.4.3 Correlation analysis

Next, we calculated the correlation coefficients between the eight conditions and the desirability of the decision and the desirability of the process in the meeting. Equally a result, a significant moderate positive correlation was found when the decision upshot was Matsuzaka beef (high risk→Matsuzaka) in the high adventure (chance emphasized) condition (r=0.57, P<.05). A significant, moderate, positive correlation (r=0.66, P<.01) was as well found when the decision issue was imported beef (depression take a chance→imported) in the low risk status. In the rule compliance condition a meaning, moderate, positive correlation was found regardless of the determination outcome (Matsuzaka beef: r=0.53, P<.05; imported beef: r=0.52, P<.05).

Read full chapter

URL:

https://www.sciencedirect.com/scientific discipline/article/pii/B9780128160329000041

Types of Variables and Measurement and Accurateness Scales

Luiz Paulo Fávero , Patrícia Belfiore , in Information Science for Business and Decision Making, 2019

two.4 Types of Variables × Number of Categories and Scales of Accurateness

Qualitative or categorical variables tin can also be classified based on the number of categories: (a) dichotomous or binary (dummies), when they only take on ii categories; (b) polychotomous, when they have on more than 2 categories.

On the other hand, metric or quantitative variables can too exist classified based on the scale of accurateness: discrete or continuous.

This nomenclature tin can exist seen in Fig. 2.11.

Fig. 2.11

Fig. 2.11. Qualitative variables × Number of categories and Quantitative variables × Scales of accuracy.

two.4.1 Dichotomous or Binary Variable (Dummy)

A dichotomous or binary variable (dummy) can merely take on two categories, and the values 0 or 1 are assigned to these categories. Value 1 is assigned when the characteristic of interest is present in the variable and value 0 if otherwise. As examples, we have: smokers (ane) and nonsmokers (0), a developed country (i) and an underdeveloped country (0), vaccinated patients (1) and nonvaccinated patients (0).

Multivariate dependence techniques have as their principal objective to specify a model that can explain and predict the behavior of i or more dependent variables through one or more explanatory variables. Many of these techniques, including the simple and multiple regression analysis, binary and multinomial logistic regression, regression for count data, and multilevel modeling, among others, can hands and coherently be applied with the use of nonmetric explanatory variables, as long as they are transformed into binary variables that represent the categories of the original qualitative variable. In this regard, a qualitative variable with n categories, for example, can exist represented past (n  −   i) binary variables.

For example, imagine a variable called Evaluation, expressed by the categories adept, average, or bad. Thus, 2 binary variables may be necessary to represent the original variable, depending on the researcher's objectives, equally shown in Table 2.7.

Table 2.7. Defining Binary Variables (Dummies) for the Variable Evaluation

Binary Variables (Dummies)
Evaluation Done D2
Good 0 0
Average 1 0
Bad 0 1

Farther details near the definition of dummy variables in confirmatory models will exist discussed in Affiliate xiii, including the presentation of the operations necessary to generate them on software such every bit Stata.

2.4.two Polychotomous Variable

A qualitative variable can take on more than two categories and, in this instance, it is chosen polychotomous. Every bit examples, we can mention social classes (lower, middle, and upper) and educational levels (elementary school, high school, college, and graduate schoolhouse).

2.iv.3 Discrete Quantitative Variable

As described in Section 2.two.two, detached quantitative variables can have on a finite set of values that ofttimes come up from a count, such as, for case, the number of children in a family unit (0, 1, 2…), the number of senators elected, or the number of cars manufactured in a certain factory.

2.four.4 Continuous Quantitative Variable

Continuous quantitative variables, on the other manus, are those whose possible values are in an interval with real numbers and result from a metric measurement, as, for example, weight, summit, or an private's salary (Bussab and Morettin, 2011).

Read total chapter

URL:

https://www.sciencedirect.com/science/article/pii/B9780128112168000021

Demographic Transition and Savings Behavior in Mauritius

Rafael Munozmoreno , ... Raja Vinesh Sannassee , in Emerging Markets and the Global Economic system, 2014

4.2 Microeconomic Modeling

4.2.ane Survey Data

We utilize data from Household Upkeep Survey (HBS) conducted in two   years 2001/2002 and 2006/2007. The HBS 2001/02 covers a sample of 6720 households, out of an estimated 300,000 private households in the country. Similarly the HBS 2006/07 surveys a sample of 6720 households, out of an estimated total 335,000 households. Each sample was selected to be representative of all households in the country through a stratified two-stage design with probability proportional to size. The survey questionnaire covers information virtually the household and household member characteristics such equally demographics, education; family size, occupation, expenditures; avails and housing conditions among others. We use the Ordinary Least Squares estimation technique for our empirical analysis.

4.ii.2 Methodology

We use a mensurate of household saving congenital on the information on income and expenditure flows provided past the HBS database. We compare the average monthly income of households and their consumption expenditures, and evaluate the part of their income that households tin save. In order to identify which factors explain household saving, nosotros guess dissimilar models. A reduced-form approach is adopted, taking into account a variety of saving determinants identified in the literature (Edwards, 1996; Loayza et al., 2000; Schmidt-Hebbel and Servén, 2000). The estimations are undertaken using Ordinary Least Squares after some robustness checks.

Model Specification

Our specification includes as dependent variable savings equally a share of income. The econometric equation is as follows:

(two) ln savincome i = α + β 1 HH i ′ + β 2 Ten i ′ + ε i ,

where the dependent variable ln savincome i is savings behavior of the caput of household (that is the ratio of savings to income), HH i ′ denotes a vector of dummies for different types of households and X i ′ is a vector including the characteristics of the household and the profile of the caput of household. ε i is a random error causeless to be independent and identically distributed. Multiple regression assay is carried out to find determinants of household savings.

The list of determinants is the monthly household income of the household caput; the gender of the household caput; age and age-squared of the household caput, the household size; the action status of the household that is whether the household is employed, unemployed, cocky-employed, or retired and the location of the household that is commune dummies.

four.2.3 Information Analysis

In this section, we clarify the income distribution and consumption pattern of all households.

Income Pattern

A comparison with poor households is also given (as per the definition of Statistics Mauritius). The income used in our analysis refers to the full household resources which comprises mainly income from employment, transfers, property, and imputed rent that is, an equivalent rental value of non-renting households. It should besides be pointed out that the income refers to the income at current prices at both 2001/02 and 2006/07 HBS. In order to let comparison over fourth dimension, we have adjusted for cost increase from 2001/02 to 2006/07.

In 2006/07, the majority (around 87%) of poor households derived a monthly income less than Rs 10,000 compared with 17% for all households. Comparing over time shows that the percentage of poor households deriving an income higher than Rs vii,500 increased from 11% in 2001/02 to 45% in 2006/07 (see Table ii).

Table two. Distribution (%) of all households by income grade, HBS 2001/02 and 2006/07

Monthly household dispensable income (Rs) 2001/02 2006/07
Households (%) Total income (%) Households (%) Full income (%)
Under 3,000 iii.5 0.five ii.i 0.2
3,000 to &lt;4,000 3.2 0.eight ane.7 0.3
4,000 to &lt;5,000 3.v 1.i ii.7 0.6
5,000 to &lt;six,000 5.0 1.9 ii.8 0.8
6,000 to &lt;vii,000 6.6 three.0 3.ix 1.3
vii,000 to &lt;viii,000 6.8 3.5 3.9 1.five
viii,000 to &lt;9,000 7.3 four.four four.seven 2.i
ix,000 to &lt;ten,000 half-dozen.7 4.v 5.1 two.v
10,000 to &lt;12,000 11.8 ix.0 10.7 half dozen.1
12,000 to &lt;14,000 9.2 8.4 9.vii six.six
xiv,000 to &lt;sixteen,000 vii.3 vii.6 ix.1 vii.2
16,000 to &lt;20,000 9.8 12.ii 12.one eleven.three
20,000 to &lt;25,000 7.6 11.8 10.5 12.three
25,000 to &lt;30,000 4.4 8.5 half dozen.5 9.four
30,000 to &lt;35,000 2.5 v.7 3.8 6.4
35,000 to &lt;40,000 1.five three.9 iii.0 5.8
twoscore,000 &amp; over 3.3 13.two 7.7 25.v

Full 100.0 100.0 100.0 100.0

Statistics Republic of mauritius, 2007

In 2006/07, the average monthly household income of poor households stood at Rs 7,055, compared with Rs 22,242 3 for all households, thus showing that the income for all households was more than than 3 times higher than that for poor households. A similar situation is observed in 2001/02. Yet, comparison of data from 2001/02 to 2006/07 shows that the average monthly household income of poor households grew by 38.9% against 33.6% for all households. Removing the upshot of change in prices over the five-yr flow, the income of poor households grew past iii.v% while that of all households dropped by 0.v% (see Tabular array iii).

Table 3. Boilerplate monthly household income (Rs) of poor households and all households, HBS 2001/02 and 2006/07

Boilerplate monthly household income Percentage increase 2001/02 to 2006/07
2001/02 2006/07 % %
Poor households 5,078 7,055 38.nine 3.5
All households 16,642 22,242 33.6 −0.5

Statistics Mauritius, 2007

Income from paid employment represented the main source of income for both poor and all households (see Table 4). The share of income from paid employment over total gross income stood at 41.0% for poor households and 59.v% for all households. Afterwards removing the issue of price changes during the v-year period, income from paid employment grew by 0.half-dozen% for poor households but dropped by 2.2% for all households.

Table 4. Boilerplate monthly household income (Rs) of poor households and all households past source of income, HBS 2001/02 and 2006/07.

2001/02 2006/07
Poor households All households Poor households All households
Paid employment 2,152 10,258 ii,906 thirteen,463
Cocky-employment 886 2,592 i,140 two,928
Transfers one,100 i,562 ane,698 2,630
Other income * 977 2,693 1,342 3,603
Average monthly gross household income v,115 17,105 seven,086 22,624
Deductions 37 463 31 382
Average monthly household income 5,078 16,642 vii,055 22,242
*
Income includes belongings income, imputed rent for non-renting households, and income from own produced goods and services.

Transfers (income from social security benefits, pension from employer, pension, allowances from parents and relatives, etc.) constituted the second main source of income for the poor. The share of transfer income over total income represented 24.0% for poor households against 11.six% for all households. Removing the event of price changes over the five-twelvemonth catamenia, transfer income grew by 15% for poor households against 25% for all households. On average, female-headed poor household earned less income than male person-headed household in both 2001/02 and 2006/07. Betwixt 2001/02 and 2006/07, income of male-headed and female-headed household increased by effectually 37% and 40%, respectively.

Expenditure Blueprint

In 2006/07, 41.7% of the poor households spent less than Rs 5000 per month compared with ix.eight% for all households. On the other hand, merely 12.0% of the poor households spent Rs ten,000 or more per month compared with 56.five% for all households. Comparison over time shows that the percentage of poor households spending at least Rs 5000 increased from 31.8% in 2001/02 to 58.3% in 2006/07. The corresponding percentage for all households increased from 79.three% to 90.ii%. Information technology should exist besides pointed out that the proportion of poor households spending Rs 10,000 or more than increased from iii.eight% to 12.0% while for all households, the corresponding per centum increased from 36.iv% to 56.5% (come across Tabular array v).

Table five. Distribution (%) of poor households and all households by consumption expenditure class, HBS 2001/02 and 2006/07

Consumption expenditure class (Rs) 2001/02 2006/07
Poor households All households Poor households All households
Below 2,500 22.8 three.8 7.3 1.3
2,500 to &lt;5,000 45.four 16.nine 34.4 viii.5
v,000 to &lt;7,500 twenty.5 23.7 31.ane 15.half-dozen
vii,500 to &lt;x,000 7.v 19.2 15.2 18.1
Total 100.0 100.0 100.0 100.0
Average monthly * four,384 x,220 half dozen,500 fourteen,300
  household
  consumption
  expenditure
*
The expenditure figures for 2001/02 have not been adjusted for infrequently purchased items such as air-tickets, household appliances, etc., while for 2006/07 an adjustment has been made.

Statistics Mauritius, 2007

In 2006/07, 41.vii% of the poor households spent less than Rs 5,000 per calendar month compared to 9.8% for all households. On the other manus, just 12.0% of the poor households spent Rs ten,000 or more per month compared with 56.5% for all households. Comparison over fourth dimension shows that the per centum of poor households spending at least Rs 5,000 increased from 31.8% in 2001/02 to 58.iii% in 2006/07. The corresponding percentage for all households increased from 79.iii% to 90.2%. It should exist also pointed out that the proportion of poor households spending Rs 10,000 or more than increased from three.8% to 12.0% while for all households, the corresponding pct increased from 36.4% to 56.five%.

From Table six below, we annotation that expenditure of households is mainly concentrated in food items and non-alcoholic beverages. Send is the next expenditure item of Mauritian households (15.2% in 2006/07 compared with 13.9% in 2001/02). This may have declined over the years with complimentary ship facilities provided to students and the elderly. Housing water, electricity, and gas too accept a high expenditure share.

Table 6. Adjusted boilerplate monthly household consumption expenditure by COICOP division—2001/02 and 2006/07 HBS

Division 2001/02 2006/07
Rs % Rs %
1. Food and not-alcoholic beverages 3,401 29.ix iv,504 29.7
2. Alcoholic beverages and tobacco 979 8.half-dozen 1,448 ix.5
3. Clothing and footwear 686 vi.0 803 v.3
4. Housing, water, electricity, gas, and other fuels one,094 ix.6 1,492 9.eight
five. Furnishing, household equipment, and routine household maintenance 909 eight.0 one,015 six.7
6. Wellness 321 2.8 466 3.ane
vii. Transport i,583 thirteen.9 ii,312 15.2
viii. Advice 359 3.1 568 3.7
9. Recreation and culture 607 5.3 759 v.0
10. Didactics 273 2.4 510 3.four
11. Restaurants and hotels 567 v.0 680 4.5
12. Miscellaneous goods and services 610 5.iv 631 4.ii

Total 11,390 100.0 15,188 100.0

Statistics Mauritius, 2007

On average, all households spent 50% more on food than poor households (Rs 4,500 against Rs 3,000). Also, the expenditure of all households on clothing and footwear, health, educational activity, and transport was around 3–five times that of poor households. Compared with all households, poor households had larger shares of their expenditure on "nutrient and non-alcoholic beverages" (46% against 32%) and "housing, water, electricity, gas, and other fuels" (15% against xi%) in 2006/07.

Household Debt

In 2006/07, the pct of indebted households, that is households having fabricated at least i loan repayment, is estimated at 46% for all households against 20% for poor households. On the average, poor indebted households disbursed Rs 1,401 per calendar month on loan repayment against Rs iv,353 for all households. The highest loan repayment for the poor households was on housing (Rs 2,491), whereas for all households the highest loan repayment was on motor vehicle (Rs 4,036) (see Tabular array 7).

Table 7. Boilerplate monthly loan repayment for poor indebted households and all indebted households by selected particular of debt, HBS 2006/07

Item of debt Poor households All households
Percentage of indebted poor households Boilerplate household debt (Rs) Percentage of indebted poor households Average household debt (Rs)
Housing 26.one 2,491 54.seven three,891
Piece of furniture 25.9 670 xiv.8 1,214
Audio and household appliances 40.ix 633 27.nine 1,133
Motor/vehicles 0.0 0 11.6 four,036
Other loan 29.8 923 40.0 2,757

Statistics Republic of mauritius, 2007

Read full chapter

URL:

https://www.sciencedirect.com/science/article/pii/B9780124115491000065