I am in agreement with Martha Pantoja's comment!

I trust it in the sense that I do not believe that the government makes anything up. I don't trust it in the sense that I am certain that there are a lot of errors, that are not intended, but that are the result of capturing everything correctly.

Economic statistics and data are always a-posteriori informed guesses.

Concerning the governmental bias, I do consider official statistical information about arms trade as unreliable.

Following

In general, a scientist has the right to doubt, including in the data provided by the state, if the method of obtaining these data is not described correctly. Moreover, to be based on a single source of data (in this case goverment) is against all laws of statistics. Maybe write: "I absolutely correctly use the data provided by an unreliable source?"

It is impossible to blame someone without evidence, but to defend based only on "faith" is no good too.

The Central Bureau of Statistics of Indonesia, is a government institution. However, it manages to keep it's independency just like the central bank, Bank Indonesia. On the other hand, the government keep itself from interfering. Both parties understand that public trust is fundamental that it needs to be handled earnestly. The government might be able to formulate a sound policy, but without public support it may never be successfully implemented.

Paul: Thank you all for answering my question. We have a good variety of countries so far.

In the case of Spain, which is my country, I would say that official statistics are very reliable in general terms.

However, some of my students do not seem to believe 100% some of the figures, paticularly those regarding unemployment and, in some cases, inflation rates.

From my point of view, the problem is that statistics are an imperfect measure of the reality... but the best one (when they are carry out correctly) and the only feasible way to do it.

What I do is to explain in greater detail to my students the underlying process to any statistics elaboration. It helps to promote confidence about statistical information among them.

I think it's a popular and up-to-date topic. I am curious about the comments coming from different countries.

No. In third world countries, we cannot rely on government statistics.

@Paul Stock Yes we do. Though government's economic statistics and data may not cent accurate. Nothing is perfect, nobody is perfect. We need believe.

I always try to find different sources of information/ data, not only national statistics, in any country I look at. In Croatia, at several cases I run to signuficant differences between different sources...

Government data must always be taken with a pinch of salt because it is heavily influenced by political considerations of successive governments. As one of the members suggested, economic analyses etc. should be based on mixed sources.

I know a lot of cases where the official data (the data provided by the government) is constantly being criticized. Official data might not be reliable, but there are some points to be made encouraging its use (I think it goes in the same way that Yuri Mansury and Eberhard Weber suggested):

- it's the data usually accepted by the multilateral ONGs and whose methodology is supposed to make it comparable with the data of other countries

- specially in middle or low income countries, there is no other agent capable of collect and manage the volume of data that the government does, so, there is no obvious substitute for it (for example, public procurement data)

- sometimes critics, who have their own spaces in mass media and influence public opinion, have their own (political) interests. They can be criticized back

When available, as Mirna Leko-Šimić mentioned, comparing among different sources is recommended (but take into account that those sources could have their own agenda and their own flaws, collecting data is a tough work).

That being said, I'd adopt this approach with a group of students: try to build a case against the data that is suspected to be unreliable or openly false. It might be a good work (it's important that it shouldn't be seen as a punishment, because it is not), and even could end up with a discussion paper or as a thesis theme. It might also be a good chance to clear some misunderstandings about what the data really means (nowadays I have discovered some misunderstandings around laboral market statistics in my country and I have written about some misusing of FDI statistics https://dolarizacion.ec/2018/02/27/inversion-extranjera-vs-inversion-extranjera-directa/ ). Finally, it might be a good chance to discuss about the theoretical framework behind the data (is the data "neutral"?).

Always governments through the economic statistics display the data depending on how they fit. If there is some improvement, for example, the unemployment rate is decreasing, many times it is emphasized. In doing so we as researchers have the task to determine the causes and factors that led to the value of such data and indicate the actual data and information obtained by their processing. Also, consideration should be given to the source of data from the data sources and the methodology of their collection, processing and presentation.

Just to add to what Kosta Sotiroski says: the media also displays the data depending on how they fit, and emphasize it as they please. In Ecuador, the case of Dinero Electrónico seems to be that the media exceeded the efforts of the government in presenting data to the public Presentation El Dinero Electrónico en el Ecuador

Paul: You make a very good point, Diego. Sometimes the media reports the data in a biased manner for their political agenda. In the USA, the TV station called CNN does this a lot.

I believe in government economic statistics, regardless of the media bias. As a researcher, I can figure out what is going on. The CNN, ABC amd FOX may show an aspect of it. But in aggregation, they are all taking about the same data. My job is to understand what is going on from the data.

Interesting!

I believe in government economic data because many sections and employees involved in procuring the relaible data

Inflation & GDP figures are acceptable and reliable. However, unemployment figures are NOT accurate in Somalia yet.

I trust my country's data, and use it in my studies and research, because I am quite sure of the validity, legality and integrity of the data, collected according to global methods, some of which may be under the supervision of international institutions.

Actually no, because there is administration corruption .

DATA today could be check, cross-checked, and verified. With the accessibility of the Internet and availability of data online from independent bodies, no government in any country could just publish any claims or data risking its own credibility and, possibly puts its own legitimacy into question. Today's IT technology and availability online data actually puts the government, if acting dishonest, on the course of behaving more honestly than if IT is not available.

In terms of research, there are alternatives. For example, working with US data there are many sources, e,g Federal reserve bank of St Louis, IMF, World Bank and European Union. The same applies to Developing countries. Chances of strong deviation of indexed values and values deviating strongly is minimal. Comparing the countries reported data to those noted above do not creates much problems. I trust data when they are available.

Do you believe your government's economic statistics and data?

Some of my college students do not trust the national government. Therefore, they don't believe the 'official' unemployment rate, inflation rate, GDP data, etc. They don't believe any economic data from the government. It is challenging to teach about these topics when some students don't trust the numbers from the government.

When you make economic analysis, you need data. Data from where? Government have some agencies on the matters (macro, labor...). You can make some surveys but it's as the additional one only.

Gove. data quality not very high and improving. OK. If you make analysis detail and use long times series, you can have some ideas on the matters. For data on labor, VN use ILO definition. You can checking...

Quality of data is low, but improving. If you don't believe its, what you use when make analysis? Gov. agencies have duty on data. It will improving under reform in Vietnam and international support...

Just trying to add some materials to this discussion:

- The Eurostat Quality on Statistics section ( https://ec.europa.eu/eurostat/statistics-explained/index.php?title=Statistics_in_development_cooperation_-_quality_in_statistics ) that highlights international commitments

- The Open Dat Watch publications about data quality ( https://opendatawatch.com/blog/indexes-of-data-quality-and-openness/ ), specially the Open Data Barometer ( http://opendatabarometer.org/ , last edition 2013 http://opendatabarometer.org/doc/1stEdition/Open-Data-Barometer-2013-Global-Report.pdf ) that highlights readiness, implementation and impact of open data provided by governments

- A collection of data quality regulations and statements (by country, region or multilateral ONG) by the UNStats ( https://unstats.un.org/unsd/methodology/dataquality/quality-references/ )

It seems this organizations are not worried about data reliability, but concerned about data availability (including the institutionality and norms to guarantee the availability of data in the long run).

I trust the government' s economic statistics and data and I am referring to them in my researches

Data sources need to be critically reviewed. But there are ranges of data sources available countries for any given indicator.

Sometimes the data is preliminary but is a benchmark, what is measured can be improved

Yes, I do believe in statistics provided by the national government of Argentina, since the whole statistics institute went through a deep re-structuring because of the fraud committed by the previous administration.

Statistics institutes follow guidelines provided by the UN, IMF and the OECD, so they have to apply well programmed protocols and are subject to control and audits. If the IMF and OECD publish and accept statistics provided by the National Institute of Statistics or office in charge, then it should suffice to prove the validity of information provided by them.

I know it is challenging to teach something if students don't trust national institutes, but I believe that assuring them that Statistics follow standardized methods and procedures should help.

Transparency, information, and providing guidelines followed by Institutes should convince students of the likelihood of data and statistics. If they keep their mistrust, there is very little you can do, since you cannot change a belief or principle.

We had for over 10 years this problem in Argentina (2005-2015) and it was really difficult to do any research, because we had to trust private sources and organizations. In that time frame, nobody believed in the information provided by government organizations, so it was really very difficult to devise any policy or do estimates of indicators and indexes.

I hope this was helpful!

Regards,

Thank you, Alejandro R. Silva-Culler . I appreciate your honest and detailed response to my question! ~ Paul

Thanks Alejandro R. Silva-Culler , it seems an interesting case the Argentinian one. Could you help us with some sources about the deep re-structuring and the previous fraud?

The data can be presented according to what you want to report and in some cases partial figures, averages, annual rates are given.

Most often that is all you have so you have to go by that hopefully in end you capture the trenfing characteristics

in participatory statistics data is generated by the community and scaled up to state level. A practical bottom-up data generation. In this case data is much more reliable and effective than it is nowadays.

Paul: Can you explain why, Naushad Khan?

No it is not correct.

Paul: What makes you think it is not correct, Fares F. Fares? Do you not trust the government's methods?

The truth and reality, the government always lies this is policy for their successful.

Even when I like the collaborative data generation (I am surprised by some results in cartography), I might not agree with Mohamed Said Samantar . In 2015 a paper argued that Wikipedia was vulnerable to information sabotage in controversial science topics Article Content Volatility of Scientific Topics in Wikipedia: A Caut...

There is some data that might be difficult to collect and sum up if it is not by governments (i.e. national accounts).

And then comes the data regulations and statements, that make the data comparable worldwide and might be difficult to achieve in a collaborative basis.

Just to be clear, I'm not saying that governments should be responsible for all data collected and available. Quite the opposite: it is a good practice to have different sources of different information that must be contrasted; I'm thinking, for example, on Panama Papers and WikiLeaks, which I can't imagine could come from an official source, but I'm also thinking on fake news and fake data that must be somehow scrutinized.

Certainly no

Best Regards Paul Stock

I will continue to use government statistics until someone produces an alternative better (but approved by government) data

What country, Osama Rahil ?

We need statistics in research papers. Where can we get them? Government statistics are more comprehensive than those that are collected for a specific purpose. There might be doubts and cynicism but what is the alternative? We can compare them across countries for verification. Organisations like IMF and World Bank have been going around showing countries how to standardize their statistics for comparable purposes. I think Govt statistics are helpful but if we are not trusting we can compile our own limited survey (might be bias) or compared with other countries statistics to have a benchmark.

For the benefit of all those who wish to have a good understanding of PARTICIPATORY STATISTICS I would invite them to go through this link:

https://www.participatorymethods.org/resource/who-counts-power-participatory-statistics

Much informative comment from Ángel Martín Román.

Unfortunately, there is no better source of information in most of the cases.

In most instances, the government data are the only available and one want hopes that measurement errors are stable over time.

Some governments pay more attention and resources in this area. What we can do is to cross-check against other countries' experience with those we are studying. A benchmark will help to put some signposts on the reliability of a particular country government number.

In individual countries, the level of trust in the government varies and is determined by many factors. If there is a high level of public distrust in the country, then the majority of citizens do not trust the statistics on various aspects of the country's economic development given by government institutions in the media. Then, conducting research using these data can be difficult. Then you should look for more reliable, objective sources of economic data. The key issue then is the level of independence of individual types and media corporations towards the government. Macroeconomic data may also come from the research departments of the central institutions of the financial system, eg from the central bank. Then the key issue is to decide if the national central bank is completely independent of the government. If so, then macroeconomic data from an independent central bank may be considered to be more objective and reliable. But this is only an example of an institution that can be independent of the government and where research is conducted on the key aspects of the development of the financial and banking system, and perhaps also the economic development of the country. Other examples of institutions that can develop government-independent reports are foreign investment banks, investment funds, rating agencies, etc. whose headquarters operate outside the country, where reports on the economic development of a specific country, particular industries and / or key industries are prepared. , financial institutions dominant in the country and enterprises from non-financial sectors.

Paul: Which country, Naushad Khan ?

Diego Fernando Ramos Flor in order to be admitted in the OECD Argentina needs to comply to their statistics standards. While I was CoS to the Secretary of Agriculture, we began the process regarding Agrifood Statistics, although the general guidelines of the Administration was to avoid providing public statistics. Then for 4 years there was a total black out, and all indexes were fabricated by altering samples and methods. There were several legal suits regarding this issue, since Bonds in Pesos were indexed by the PCI. The actual administration of the INDEC has been audited by the OECD, and a delegation of the IMF was sent to check procedures and standards. Check for the year 2016 for re structuring and 2007 for the institutional irregularities. https://www.indec.gob.ar/indec/web/Institucional-Indec-Historia

Diego Fernando Ramos Flor Regarding the OECD Mission that came in 2016. http://www.minutodecierre.com/nota/2016-9-21-para-ingresar-a-la-ocde-el-gobierno-primero-debe-mostrar-un-indec-renovado

In many cases data coul be "Soften" with average, for example percapita income.

No, I cant beleive.

Paul: What country, Muhammad Abdullah Avais ?

Why can't you believe the government data?

Some times data is estimated, because is difficult and costly to get all information.

You have to believe because you don’t have other choices and if when other choices available they may not be cite able. Cite ability is very important for authenticity

Government information and data are needed as given by governing authorities because they have greater capacity than an individual to collect the data. Indeed more believable by the general public than one produced by one person

Measurement errors that are if they continue over time are usually non-stationary and would not affect the final outcome in the long run

The question is: Do we have alternative source of data? Collecting our own primary data is time consuming and costly.

Akchmad,

Precisely because of the costly is that for example in the case of Mexico, government institutions are responsible for data collection.

When it is the only available data, we have to use what is available or we have to do without

For my country, definitely NOT 😔 - Iraq

Following answers

No

Paul: Please tell us what country you are from when answering. Thank you!

In economic history, we have seen countries with money from oil/mineral resources but not properly utilized are also wasted. Their short term spurge of money did not bring benefit to the people which lack education, institutions and rule of law. Corruption abounds and wastage is high. On the other hand, there are countries with little natural resources but an educated labor force which is able to harness whatever they don't have from the rest of the world to grow efficiently and free trade.

Paul: Can you please give an example of each type of country, Chung Tin Fah ?

One of the objectives of the information is that it allows decisions.

The issue is not one of believe but it is what you are supposed to believe in the absence of other sources

To avoid having different data on the economy, finance, exchange rate, external debt, among others, only government and private institutions are empowered to provide data.