Week 1



1.It always starts with hypotheses.
2.Significance tests are always an esential element of statistical analysis.
3.The level of association or correlation between variables is enough for explaining studied
phenomena.

ARE THESE MYTHS TRUE???

}NO. NOT ALWAYS
}Every research should have a problem, which is transferred into a researchable form by a research
question.
}There are certain forms of RQs.
}Only question WHY leads scholars to researches that are based on theory and thus require
hypotheses.
}There are 2 types of hypotheses – theoretical and statistical

}NO. THEY AREN’T!!!
}Significance tests are part of the inferential statistics.
}Inferential statistics can be used only when we operate with probability samples.
}Inferential statistics = generalize to the whole population – to be discussed later
}These tests cannot be used when operating with data from census or quota, specific, snowball
samples.

}NO, IT IS NOT!!!
}Meaurement of association and correlation is a higher level of description.
}Although this association can help to understand the phenomenon, they cannot bring answer to
question „why“ because why implies finding the causes behind it.
}However, if we want to explain a phenomenon, first we have to find the association – where there
is no probabilistic association, there is not casual link.

}Gathered data ` research design ` research question óexplain a problem
}
}Standardized data gathered by use of questionnaire  and samples.
}
}Unit of analysis = individual, groups,…

}To find out the status of the individual characteristics the dataset is based on, to identify
their characterstics, to describe them based on the characteristics of the dataset,  to identify
the distribution of them in the sample
}To find the causes of variabilty or the relations or links among various characteristics
}To monitor the trends/changes in the quantified charactersitics of the dataset by use of
longitudinal survey

}to present and interpret numerical data
}
}to determine the relationship between one thing (an independent variable) and another (a dependent
or outcome variable) in a population
}
}

}Use of quantitative design
}
}Large samples
}
}Use of representative samples
}
}Collection of data in quantifiable form

}What are “statistics”?
}
}
}A set of mathematics concerned with understanding and summarizing collections of numbers
}A collection of numerical facts
}Estimates of population parameters ð derived from samples
}

}Descriptives statistics
}
}
}Inferential statistics

}Unit of analysis to be studied – how we choose them a sample design (probabilistic, quota,
deliberate, etc.)
}Gathering data by observation and measurement
}Controling data a cleaning data
}Gathering first basic information on the dataset a univariate analysis
}Calculating statistics and monitoring time and content changes
}Stating the distribution of the phenomena, relational or correlation analysis
}Running statistical checking of the hypotheses
}Running inferential statistics
}Running multilevel analysis

}We have to settle the general universe or population.
}Criteria for settling our general population are given by our research topic.
}We are very rarely able to study the whole general population.
}BUT if we do it, we operate with Census
}Usually we operate with samples and sample surveys
}Sample should be representative for our population.

}Selected sample has the same/analogous structure as the general population from the point of view
of various known and unknown characteristics ó probability sample
}
}
}Results can be generalized to the whole general population from where the sample was drawn.
}
}

}Aim to characterize and make inferences not about a particular person but about all people or all
people having certain characteristics.
}These groups of interest are called populations.
}Typically, these populations are too large and inaccessible to study.
}Instead, we study a subset of the group, called a sample.
}In order to make reliable inferences about the population, samples are ideally randomly selected.
}The population properties of interest are called parameters.
}The corresponding measurements made on our samples are called statistics.

}When measuring a certain phenomenon we need to know how it is delimited, defined.
}We need its concept and definition!!!
}We measure different aspects of the phenomena:
1.Intensity of the characteristic (both for the units of analysis and the external context)
2.Distance of objects
3.Dependece (asymetrical relation) and relationship (symetrical relation)
4.Global characteristics of the sample

}Constants are properties that never change
}
}Most sociological parameters vary considerably
◦Between individuals (e.g., political preference)
◦Within individuals (e.g., heart rate)
}
}Any variable whose variation is somewhat unpredictable is called a random variable
}

}
}Categorical data
}There are basically two kinds of data in this
}groups:
}
}Nominal data
}Ordinal data
}
}Continuous data is (sometime) referred to as interval data.
}

}Interval scale:  values are numerically meaningful, and interval between two values is meaningful.
}
}Ratio scale:  opertes with a 0 value
}
}Dichotomous variables: yes/no, belong/not belong response scale
◦

www.socialresearchmethods.net



}A continuous variable may assume any real value within some range
}Total fertility rate 1960-2006


}A discrete variable may assume only a countable number of values (intermediate values are not
meaningful).
}The number of children ever born to women aged 40-71 in the CR
}(1991 and 2001)

}Studies involve independent and dependent variables.
◦
◦1. The independent variable is controlled by the researcher.
◦2. The dependent variable is measured.
◦
nWe seek to detect and model effects of the independent variable on the dependent variable.
◦
◦How else are they named?

Predicted and outcome

}A predicted or expected answer to a research question
}
}Example: Education is positively correlated to salary

}We call the hypothesis that you support (that supports your prediction) the alternative
hypothesis,
}We call the hypothesis that describes the remaining possible outcomes the null hypothesis.
}Sometimes we use a notation like HA or H1 to represent the alternative hypothesis or your
prediction, and HO or H0 to represent the null case.
}
}One-tail vs. Two-tails hypotheses

}SPSS = Statistical Package for the Social Sciences
}Was developed in the late 1960’s by American political scientists (Norman Nie)
}Is not the best or most advanced statistical software, but combines comprehensiveness and
accessibility
}Still widely used in academic and commercial research
}

}to introduce the basic notions in statistics - population, parameters, descriptive statistics,
inferential statistics, sample, variables, etc.
}to build and use variables and databases using SPSS
}to introduce the basic statistical data analysis methods using SPSS

}Field, A. - Discovering statistics using SPSS for Windows : advanced techniques for the beginner,
London: Sage Publications, 2009
}Vaus, D.A. de - Analyzing Social Science Data, London: Sage Publications, 2002
}Hardy, M. A; Bryman, A. Handbook of data analysis, London : Sage Publications, 2004
}Miller, R.L. - SPSS for Social Scientists, Houndsmill: Palgrave, 2002

}European Values Survey dataset – Wave 3 – 2008 – 3 countries: Czech Republic, Finland and France
}European Social Survey – ESS Cumulative Dataset Rouds 1-7 (data for 4 countries: Austria, Czech
Republic, Finland and France)
}

}1st part a lecture
}
}2nd part a seminar – individual work on PC
}
}

}
}
}1st term
}
}2nd term